Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball [1st ed.] 978-3-030-17279-4;978-3-030-17280-0

Table of contents :
Front Matter ....Pages i-xxvi
Front Matter ....Pages 1-1
The Roots of Baseball in the US (Jeffrey S. Markowitz)....Pages 3-11
A Mini-history of Latin America and Professional Baseball (Jeffrey S. Markowitz)....Pages 13-19
Literature and Statistical Review of Race/Hispanic Origin and Mortality (Jeffrey S. Markowitz)....Pages 21-29
Front Matter ....Pages 31-31
Study Methods (Jeffrey S. Markowitz)....Pages 33-44
Births and Deaths in the Study Cohort (Jeffrey S. Markowitz)....Pages 45-48
Preliminary Testing of Main Study Hypotheses (Jeffrey S. Markowitz)....Pages 49-51
Front Matter ....Pages 53-53
The Role of Educational Attainment (Jeffrey S. Markowitz)....Pages 55-61
Examination of Other Independent Variables (Jeffrey S. Markowitz)....Pages 63-84
Final Multivariate Testing of Study Hypotheses (Jeffrey S. Markowitz)....Pages 85-90
Front Matter ....Pages 91-91
Comparison of Mortality Rates Between Major League Baseball Players and the General Population (Jeffrey S. Markowitz)....Pages 93-99
Summary, Conclusions, and Implications (Jeffrey S. Markowitz)....Pages 101-116
Back Matter ....Pages 117-118

Citation preview

SPRINGER BRIEFS IN PUBLIC HEALTH

Jeffrey S. Markowitz

Mortality Among Hispanic and AfricanAmerican Players After Desegregation in Major League Baseball

SpringerBriefs in Public Health

SpringerBriefs in Public Health present concise summaries of cutting-edge research and practical applications from across the entire field of public health, with contributions from medicine, bioethics, health economics, public policy, biostatistics, and sociology. The focus of the series is to highlight current topics in public health of interest to a global audience, including health care policy; social determinants of health; health issues in developing countries; new research methods; chronic and infectious disease epidemics; and innovative health interventions. Featuring compact volumes of 50 to 125 pages, the series covers a range of content from professional to academic. Possible volumes in the series may consist of timely reports of state-of-the art analytical techniques, reports from the field, snapshots of hot and/or emerging topics, elaborated theses, literature reviews, and in-depth case studies. Both solicited and unsolicited manuscripts are considered for publication in this series. Briefs are published as part of Springer’s eBook collection, with millions of users worldwide. In addition, Briefs are available for individual print and electronic purchase. Briefs are characterized by fast, global electronic dissemination, standard publishing contracts, easy-to-use manuscript preparation and formatting guidelines, and expedited production schedules. We aim for publication 8–12 weeks after acceptance.

More information about this series at http://www.springer.com/series/10138

Jeffrey S. Markowitz

Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball

123

Jeffrey S. Markowitz Health Data Analytics Princeton Junction, NJ, USA

ISSN 2192-3698 ISSN 2192-3701 (electronic) SpringerBriefs in Public Health ISBN 978-3-030-17279-4 ISBN 978-3-030-17280-0 (eBook) https://doi.org/10.1007/978-3-030-17280-0 © The Author(s), under exclusive license to Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

Growing up in an Epicenter In the mid-1950s, I was a young boy growing up in Brooklyn, less than 5 miles from Ebbets Field, the home of the Brooklyn Dodgers. The train ride to Ebbets Field cost 15 cents and took about 20 min. As I would understand years later, I was growing up near the epicenter of baseball’s “Great Experiment” (Tygiel 2008) of reintegrating African-Americans into Major League Baseball after more than a half-century hiatus. Known as “Dem Bums,” a nickname derived from a complicated mix of fans’ affection and derision, most Brooklynites, like myself and my friends, loved the Dodgers. As I saw things, even at my youthful age, there were two major obstacles constantly confronting Dem Bums: the New York Giants, the inter-borough, intra-league rivals who played at the Polo Grounds in upper Manhattan, and the New York Yankees, the “Bombers” from the Bronx. Most of us fiercely disliked the Giants as they had a knack of preventing the Dodgers from getting into the World Series. Perhaps, the best-known example of this occurred in 1951 when the Giants beat the Dodgers on a home run in the bottom of the ninth inning in the final game of their National League playoff series to advance to the World Series. Russ Hodges’ famous radio call of the home run hit by Bobby Thompson—also known as “the shot heard round the world”—that won the game and the National League pennant for the Giants that afternoon added insult to injury for Dodger fans. With uncontrollable glee, Hodges shouted repeatedly: “the Giants win the pennant, the Giants win the pennant….” a total of 5 times. Though too young to recall this tragic moment, I have encountered stark reminders hundreds of times throughout my life. As much as my friends and I disliked the Giants, it was the New York Yankees that most Dodger fans truly despised. Playing their home games at Yankee Stadium, the so-called “house that [Babe] Ruth built,” the Yankees were the gold standard in baseball. Between 1921 and 1959, the Yankees appeared in 24 World Series and won 18 of them. In comparison, the Dodgers played in a total of eight World Series during this same period and won just two of these. And guess who beat the Dodgers

v

vi

Preface

in these six World Series losses? If you guessed the Yankees, unfortunately, you are correct. It was an intense, lopsided, bitter, intra-city rivalry that New York City baseball fans cherished year after year. While the Yankees dominated the Dodgers with respect to winning and losing baseball games, it was the Dodgers who led the way in desegregating Major League Baseball. Of course, this wasn’t a competition or anything, but in 1947, Jackie Robinson, playing for the Brooklyn Dodgers, became the first African-American to play in the Major Leagues since before the turn of the twentieth century. (It took the Yankees another 8 years to field their first African-American player, Elston Howard, in 1955.) To this day, I’m proud that the Dodgers were the first team to end the ugly scourge that was the Major League Baseball’s color barrier. Putting aside “beating the Yankees” at something, the signing of Jackie Robinson by the Dodgers was a great source of pride to me and many others living in Brooklyn. Jackie Robinson proved to be an outstanding baseball player during his 10-year Major League career, leading the Dodgers to their first World Series victory ever— beating the Yankees in 1955. Of far more importance, Robinson and the end of segregation in Major League Baseball helped to change the course of US race relations, and even US history. Several writers and historians believe that Robinson’s entry into Major League Baseball served as the impetus for the US Civil Rights movement in the 1960s (Constitutional Rights Foundation 2018; Craig 2017). My own love for baseball and its history was rooted in these Brooklyn years and has endured throughout my life.

My Uncles’ Influence Two uncles on my mother’s side of the family, also Brooklyn residents, often came together to visit our family. During these visits, all my uncles wanted to do was watch baseball on our tiny black and white 21-inch TV. Though not many games were televised in the mid-1950s, my uncles always managed to appear for visits on days when there were games on TV, and exactly at the time when the first pitch would be thrown. Like my parents, both uncles struggled to make ends meet. In my parents’ case, financial woes left them with little time or energy to develop an interest in baseball. My uncles clearly adored the Dodgers, but also showed considerable interest in several non-Dodger players of the time like Roberto Clemente, Orlando Cepeda, Luis Aparicio, Minnie Minoso, and others. As I would learn years later, these were some of greatest baseball players of all-time, all born in Latin American countries. But why did my uncles care so much about baseball, in general, and these players specifically? As the years passed, and I better understood my uncles’ backgrounds, I began to appreciate their fervor for baseball and their special interest in Latin American players. Both uncles had been born and raised in Havana, Cuba during

Preface

vii

the 1920s and 1930s where baseball had become very popular. (Both my parents also grew up in Cuba but were born in Poland.) While baseball originated in the US, Cuba was the first Latin American country where the game was played. Baseball had become part of Cuba’s national identity as they fought to gain their independence from Spain during the latter part of the nineteenth century. For many Cubans, baseball was a nationalistic treasure, an alternative to Spanish bullfighting, and a valuable part of their newly found independence. It was the Cubans who organized the first Latin American baseball league, and who helped spread baseball to other Latin American countries. To this day, baseball remains an integral part of Cuban life and culture, and like the US, baseball, rather than soccer, is Cuba’s proud national pastime. During the first half of the twentieth century, when African-Americans were banned from baseball, several Latin American players appeared on Major League Baseball rosters. Not only did these players have to be very talented, but their skin color had to be something other than black. Coinciding with the end of baseball’s color line in 1947, Latin American players, even ones with dark skin color and visible African ancestry, began appearing in larger numbers on Major League Baseball rosters.

Curious About the Past Baseball was, and still is, a sociocultural phenomenon with a complicated history. Combined with my uncles’ influence, growing up at “Ground Zero” of baseball’s integration piqued an evolving curiosity about the game and its players. This interest grew stronger as I progressed through high school, undergraduate, and graduate school. Growing up in a Northeastern US city that was supposed to be relatively tolerant of racial and ethnic minorities, I can assure you that African-Americans as well as Hispanics faced racism and discrimination on and off the baseball field. I witnessed this on the streets of Brooklyn as well as at Dodger games. I saw first-hand how Jackie Robinson and other African-American players that followed him into the League were treated by some white players, coaches, and fans. I saw the bean-balling and the spiking and heard the racial epithets. These unfortunate realities of that time have been documented in more than a few books (Zeiler 2013; Ruck 2011; Moore 2011; Lamb 2012; Tygiel 2008; Rust Jr 1992; Burgos Jr 2007). “The larger history—of racial struggle in Brooklyn and America after World War II —is often ugly and painful” (Sokol 2015). As a Public Health student at Columbia University in New York City in the 1980s, I asked myself questions about race, national origin, and how this could impact on health. I knew that baseball was a statistics-crazy sport and believed that it might be possible to collect data on this subject. However, there was no Internet then and the available data on players’ health and mortality was scanty. But now, nearly 40 years later, with an enormous amount of verifiable data on the Internet,

viii

Preface

the aging of players from the time of integration, plus decades of personal experience as an epidemiologist, I’ve decided to return to earlier questions from my youth and study the connection between race/Hispanic origin and mortality among former Major League Baseball players—specifically, by studying former players who played during the 40-year period after segregation had ended in the Majors.

Studying Death and Its Fairness For many, the concept of death may be difficult to consider and I’m no exception. It’s been painful to examine the deaths of many of my childhood heroes. Before writing this book, my previous (third) book titled “Mortality and its risk factors among professional athletes: A comparison between former NBA and NFL players” (Markowitz 2018) focused on comparing mortality risk between two professional groups of former players. Altogether, I’ve been studying mortality for about 5–6 years now. The death of most professional athletes, including Major League Baseball players, is generally a public matter, particularly when a player is a star, dies young, or has a nonnatural cause of death. Despite the certainty of death, hearing about the passing away of a former player always surprises me. It’s hard to fathom how strong, remarkable athletes can succumb to anything—even death. Nevertheless, the study of death can tell us much about life, perhaps even help to extend life in a quality manner for future generations. The study of special groups, like professional ballplayers, who may differ from their general population counterparts in terms of income, body size, and athleticism may also provide important preventative clues. Consequently, the study of death is indispensable to public health. Hence, while saddened when hearing about a players’ passing, I feel fortunate to have the opportunity to study mortality among former ballplayers, many of whom I watched play and admired for years. Several things about death that make it amenable for epidemiological study is that, with advancing age especially, it occurs in relatively large numbers, and there’s variability with respect to when it occurs. In addition, while sometimes cruel and unexpected, there’s nothing random about death from a statistical perspective. Readers should rest assured that this is not a philosophy book, but a fundamental question related to mortality has to do with its fairness. Putting aside issues related to pain and suffering, I think most would agree that death affecting the elderly disproportionately more than the young seems “fair.” There are also gender differences in life expectancy. As the “life expectancy champions” (Zarulli et al. 2017), females currently live longer than males in the US as well as almost every place else in the world (Barford et al. 2006). This is largely due to biological, genetic, social, and environmental factors. This may seem unfair (especially if you’re a guy), but since it’s a worldwide phenomenon and the factors advantaging females are largely immutable, many (including myself) will accept their longer longevity as being fair.

Preface

ix

However, is it equitable that whites in the US tend to live longer than African-Americans? When one considers the possible man-made reasons for these differences, such as education, income, access to quality health care, and discrimination, it becomes hard to justify the fairness of between-race mortality differences that advantage one group over another. Moreover, many public health and government officials agree with me on this matter. In 1990, for example, the US Department of Health and Human Services issued its Healthy People 2000 initiative (CDC 2009), which promoted a comprehensive strategic plan for improving health in the US by the year 2000. This plan contained hundreds of health-related goals and objectives including the need “… to address the overreaching goal of reducing health disparities in special populations at higher risk than the total population for death, disease, or disability” (National Center for Health Statistics 2001). Some specific objectives of the plan focused on reducing mortality-related disparities that disadvantaged African-Americans. By 2010, the goals of the Healthy People 2020 initiative focused on eliminating rather than reducing health disparities, and for 2020, 1 of 4 overarching goals of Healthy People 2030 was to “achieve health equity, eliminate disparities, and improve the health of all groups” (CDC 2011). The sad fact is that mortality differences based on race have been documented in the US general population since records have been kept. But do such disparities have to continue, and what can be learned from special groups like former professional athletes to help reduce or eliminate between-race mortality differences? Does Hispanic origin play a role in mortality risk among former Major League Baseball players including those who played during the post-segregation era? These and other important questions will be addressed in this book.

Brief Overview of this Book This book studies mortality and mortality risk among more than 5200 former Major League Baseball players. Individuals who participated in the league as players anytime between 1947, the year that baseball desegregated, and 1986 will be studied. Most of this cohort, about 80%, is non-Hispanic whites, and certainly there are many great men and outstanding baseball players within this group. Without diminishing the contributions to the game made by former non-Hispanic white players, the significance of the roughly 1000 African-American and Hispanic individuals who played Major League Baseball during the post-segregation years cannot be overstated. By all accounts, many of these players endured ridicule, scorn, and other indignities from fans, the media, playing opponents, and teammates. Many African-Americans were subject to discrimination before, during and even after their baseball playing careers had concluded. In many ways, Major League Baseball was (and may still be) a reflection of the broader US culture which in some circles embraces racism as a way of life.

x

Preface

Many of the African-American and Latin American players who played Major League Baseball following the end of segregation were trailblazers and heroes with numerous books and other publications describing how these players overcame poverty, little or no education, difficult living situations and other adverse circumstances to play professional baseball (Zeiler 2013; Ruck 2011; Moore 2011; Lamb 2012; Tygiel 2008; Rust Jr 1992; Burgos Jr 2007). With the passage of time, the aging of these 1000 African-American and Hispanic players has now also made it possible to examine another important part of their story—their mortality. Some individuals who played Major League Baseball in the late 1940s could have been born during the first decade of the twentieth century, and, predictably, most of these players have now passed away. In addition, players born in the 1920s (and later) may have already passed away too. Given the roughly 1000 African-American and Latin American players who participated in Major League Baseball during the post-segregation period (up to the year 1986) and their advancing ages as of 2018, it has now become feasible to statistically analyze comparative mortality among these groups of players based on their race and national origin. While the vital status of these African-American and Hispanic players is the focus of this book, there were also more than 4200 non-Hispanic white players who played Major League Baseball between 1947 and 1986 who are obviously a vital part of the story as well, and their mortality statistics become critical for purposes of comparison. Before delving into the empirical study that’s the focus of this book, it’s essential to examine the history and evolution of professional baseball in both the US and Latin America. Chapter 1 contains a brief history of baseball as well as its social and cultural underpinnings in the US. The roots of baseball cannot be fully separated from several critical historical events in the US like the Civil War and Reconstruction. The end of the Civil War led to the abolition of slavery. Yet despite Constitutional amendments passed between 1865 and 1870 that were supposed to help equalize rights between the races, discrimination and racism persisted. Jim Crow laws and court rulings during the first part of twentieth century continued to foster an uneven culture that favored whites over African-Americans, particularly those residing in the Southern part of the US. Chapter 2 offers a very brief history of Latin America that is centered on those countries that have contributed players to Major League Baseball during the years following the end of segregation. Like the US, Latin America was colonized by Europeans who sought to use slave labor as the basis of their economies. Before the slavery would end, a very racially and ethnically mixed culture would evolve in Latin America comprised of various combinations of indigenous people, Europeans and Africans. Baseball, essentially an import from the US, first arrived in Cuba and then became popular in several other Latin American countries. For more than a century now, US general population mortality statistics consistently indicate that African-Americans are disadvantaged mortality-wise compared to whites. The literature and statistical review described in Chap. 3 documents these inequalities based on data published by the Centers for Disease Control and Prevention (CDC). Not only is there a long-standing gap in life

Preface

xi

expectancy between the races, but overall death rates as well as death rates for heart disease and cancer are consistently higher among African-Americans. The CDC statistics also indicate that Hispanics residing in the US have better mortality statistics than African-Americans, and often fare better than non-Hispanic whites. However, statistics can sometimes be complicated and this “Hispanic Paradox”— the idea that Hispanics living in the US tend to have equal or better mortality outcomes relative to non-Hispanic whites despite being disadvantaged socioeconomically—warrants closer scrutiny. The empirical analysis begins in Chap. 4 with a description of the methods and the study hypotheses. Based on the literature review provided in Chap. 3, many readers will anticipate mortality predictions for African-American players relative to their non-Hispanic white counterparts. In contrast, given the uncertain state of mortality patterns among Hispanics, hypotheses related to comparisons between former Major League Baseball players from Latin American countries will be more difficult to formulate. Chapter 5 presents basic descriptive statistics for key study variables like race/Hispanic origin, year of birth, and vital status. Simple cross-tabulations of vital status by race/Hispanic origin are given that are stratified by year of birth categories. Very preliminary testing of the study hypotheses will be described in Chap. 6. The goal of the analyses described in Chaps. 7 and 8 will be to identify other variables that could be important in understanding the relationship between race/Hispanic origin and mortality within the study cohort. As will be shown, there are several other variables besides race/Hispanic origin—such as educational attainment (Chap. 7), body mass index (Chap. 8, Sect. 8.2), US birthplace region (Chap. 8, Sect. 8.4) number of career years played Major League Baseball (Chap. 8, Sect. 8.6), player position (Chap. 8, Sect. 8.8), and handedness (Chap. 8, Sect. 8.10) —that may be statistically related to either or both race/Hispanic origin as well as mortality. The question then becomes whether these items add independently to the prediction of mortality as risk or protective factors, and/or change the statistical association between race/Hispanic origin and mortality. Multivariate independent predictors of mortality will be uncovered in Chap. 9. Within the boundaries of the variables included in this study, this multivariate analysis will serve as the definitive test of whether race/Hispanic origin is a risk factor for mortality among the former Major League Baseball players who played in the era following desegregation. Up to this point in the book, all the empirical analyses are “internal” since they are restricted to players in the study cohort. Chapter 10 is an “external” analysis because it compares mortality rates observed among cohort players to the US general population, matched for age, gender, and race. Standardized Mortality Ratios (SMRs) will be generated for each race/Hispanic origin group using US Census data answering the question whether mortality rates are higher, lower, or about the same between the three groups of professional athletes and their general population counterparts. Chapter 11 will offer my perspective on what the study findings mean and where we go from here in terms of future research needs. Study limitations will also be discussed.

xii

Preface

Delving into race and mortality disparities that exist more broadly in the US general population will not be the easiest or most comfortable matter to discuss. Yet, in the context of this pressing public health issue, a dialogue is needed to further unravel the causes of disparities and what can be done to reduce them.

Ground Rules Use of Player Names Player names will be used sparingly and only in chapters that do not contain empirical analyses. As many baseball and other sports fans know, there’s a wealth of demographic and player-related information published online and in hundreds of sports books and publications. Nevertheless, details related to player lives and deaths could be sensitive material for surviving family members, friends, and fans, especially when this information is presented in complex ways involving combinations of variables that, heretofore, have not been made public. Out of respect for research privacy rules and policies like those published by HIPAA, or Health Insurance Portability and Accounting, (US Department of Health and Human Services 2003) player names will not be included within the empirical chapters of this book. References to Race and Hispanic Origin The main independent variable in this study is race/Hispanic origin. Players born in the US are divided into two racial groups, African-Americans and non-Hispanic whites. The third group consists of players born in Latin America. I have chosen to use the term “African-Americans” in this book. Other authors quoted in this book have used other terms and expressions which may include “Blacks,” “Black Americans,” and “Afro-Americans.” In some cases, these terms can be used interchangeably, although others have different meanings. In my writing, I attempt to help readers better understand what is meant by other authors when they use different words to describe various racial and ethnic groups. My decision to use “Hispanics” rather than “Latinos” was a difficult one to make since the use of both words has pros and cons. The main factor for me in deciding to use “Hispanics” is that it’s used by many others cited in this book including government organizations and authors. For example, I often sourced general population data on “Hispanics” published by the CDC and refer to the “Hispanic Paradox” dozens of times. Although I prefer Latino as the more empowering term from a cultural and historical perspective, using Hispanic is more consistent with the relevant literature. For purposes of this book, Major League Baseball players born in the US who are not African-American are white. Moreover, because we also know that they were not born in a Latin American country, these players will be called “non-Hispanic whites.” Based on an analysis of Hispanic surnames (Word and Perkins 1996), there are only about a dozen US players in the cohort who come

Preface

xiii

from Hispanic backgrounds, and while not perfect, these players will be aggregated with the non-Hispanic white players. To summarize, the 3 groups that are the focus of this book are African-Americans, non-Hispanic whites, and Hispanics. The name I’ll use to describe this grouping variable will be “race/Hispanic origin.” Other authors may use alternative words and terms to describe various races or other combinations of race and national origin. When this occurs, the authors’ words and terms will be noted and used in the text and table displays. The Need for Solid Statistics This is a research study that examines specific hypotheses related to how mortality risk may be related to the race/Hispanic origin of former Major League Baseball players. In addition, strategies are used to rule out alternative explanations. Consequently, the approach is based on sound statistical methods and techniques. While many readers may be interested in the subject matters covered in this book, some will not have the statistical background necessary to understand the analyses presented in Chaps. 6–10. In these chapters, an attempt is made to explain statistical matters to accommodate a broad range of readers. To this end, each empirical chapter concludes with a “Summary of Empirical Results” section that will detail, in nontechnical terms, the statistical conclusions that have been reached. Also, readers interested in nonstatistical explanations of results should look for sentences that begin with the phrases “In other words….” or “This means…” It may be difficult to “enjoy” a book that focuses on mortality. This is particularly true when the deceased includes individuals who many see as sports heroes. So instead of saying that I hope readers enjoy this book, my goal is that you find it interesting, informative, and meaningful. As a public health professional, my primary objective is to help extend quantity and improve quality of life, particularly for those with the greatest need. Princeton Junction, NJ, USA

Jeffrey S. Markowitz

References Barford, A., Dorling, D., Davey Smith, G., & Shaw, M. (2006). Life expectancy: Women now on top everywhere. BMJ, 332(7545), 808. Burgos A., Jr. (2007). Playing America’s game: Baseball, Latinos, and the color line. Berkeley: University of California Press. CDC, Centers for Disease Control and Prevention. (2009). Healthy People 2000. Retrieved from https://www.cdc.gov/nchs/healthy_people/hp2000.htm. CDC, Centers for Disease Control and Prevention. (2011). Healthy People 2020. Retrieved from https://www.cdc.gov/nchs/healthy_people/hp2020.htm. Constitutional Rights Foundation. (2018). Jackie Robinson: Desegregation begins with baseball. Retrieved from http://www.crf-usa.org/black-history-month/jackie-robinson. Craig, M. (2017). Baseball’s integration and the contemporary civil rights movement. Retrieved from https://www.beyondtheboxscore.com/2017/9/25/16356568/jackie-robinson-integrationcivil-rights-movement-jesse-owens-sean-doolittle-bruce-maxwell-racism.

xiv

Preface

National Center for Health Statistics. (2001). Healthy People 2000 Final Review, Hyattsville, Public Health Service. Retrieved from https://www.cdc.gov/nchs/data/hp2000/hp2k01.pdf. Lamb, C. (2012). Conspiracy of silence: Sportswriters and the long campaign to desegregate baseball. Omaha: University of Nebraska Press. Markowitz, J. S. (2018). Mortality and its risk factors among professional athletes: A comparison between former NBA and NFL players. Cham: Springer Nature. Moore, J. T. (1988). Larry Doby: The struggle of the American League’s first Black player. Mineola: Dover Publications, Inc. Ruck, R. (2011). Raceball: How the Major Leagues colonized the Black and Latin game. Boston: Beacon Press. Rust A., Jr. (1992). Get that Nigger off the field: An oral history of Black ballplayers from the Negro Leagues to the present. New York: Delacorte Press. Sokol, J. (2015). Jackie Robinson’s life was no home run for racial progress. Time. Retrieved from http://time.com/3942084/jackie-robinson-racial-progress/. Tygiel, J. (2008). Baseball’s great experiment. New York: Oxford University Press. US Department of Health and Human Services. (2003). OCR HIPAA Privacy, Research. Retrieved from https://www.hhs.gov/sites/default/files/ocr/privacy/hipaa/understanding/special/research/ research.pdf. Word, D. L., & Perkins C., Jr. (1996). Building a Spanish surname list for the 1990s—a new approach to an old problem, Technical Working Paper No. 13. Retrieved from https://www. census.gov/population/documentation/twpno13.pdf. Zarulli, V., Barthold Jones, J. A., Oksuzyan, A., Lindahl-Jacobsen, R., Christensen, K., & Vaupel, J. W. (2017). Women live longer than men even during severe famines and epidemics. Retrieved from http://www.pnas.org/content/pnas/early/2018/01/03/1701535115.full.pdf. Zeiler, T. W. (2013). Jackie Robinson and race in America. Boston: Bedford/St. Martin’s.

Acknowledgments

Many thanks to my dear wife and long-time academic partner, Dr. Elane Gutterman, Ph.D., for her outstanding editing as well as her many thoughtful suggestions for the content of this book.

xv

Contents

Part I

The Backstory . . . . . . . . .

3 3 4 5 6 7 8 9 10

..... .....

13 13

. . . . . . . .

. . . . . . . .

14 15 15 16 16 17 17 17

..... .....

18 18

Literature and Statistical Review of Race/Hispanic Origin and Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 General Population Mortality Statistics . . . . . . . . . . . . . . . . . . . 3.2 CDC General Population Mortality-Related Statistics . . . . . . . .

21 21 22

1

The Roots of Baseball in the US . . . . . . . . . . . . . . . . . . 1.1 Baseball in the US . . . . . . . . . . . . . . . . . . . . . . . . 1.2 African-Americans and Slavery in the US . . . . . . . 1.3 African-Americans in Baseball . . . . . . . . . . . . . . . . 1.4 Baseball and the Larger American Society . . . . . . . 1.5 The Role of World War II . . . . . . . . . . . . . . . . . . . 1.6 Desegregation in Major League Baseball . . . . . . . . 1.7 Jackie Robinson and the Desegregation of Baseball References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2

A Mini-history of Latin America and Professional Baseball . 2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Race/Ethnicity and Discrimination in Latin America Versus the US . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Hispanics in Baseball . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Cuba . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Venezuela . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.4 Puerto Rico . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.5 Dominican Republic . . . . . . . . . . . . . . . . . . . . . 2.3.6 Panama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Journeys Travelled by Hispanic Players to Major League Baseball . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . .

. . . . . . . . .

. . . . . . . .

. . . . . . . . .

. . . . . . . .

xvii

xviii

Contents

3.3 3.4 3.5 3.6 3.7

Life Expectancy . . . . . . . . . . . . . . . . . All-Cause Death Rates . . . . . . . . . . . . Heart Disease and Cancer Death Rates . The Hispanic Paradox . . . . . . . . . . . . . Professional Athlete Literature . . . . . . . 3.7.1 Baseball . . . . . . . . . . . . . . . . . 3.7.2 Basketball . . . . . . . . . . . . . . . 3.7.3 Football . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

22 23 24 25 26 27 27 27 28

Study Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Study Objective and Design . . . . . . . . . . . . . . . . . . . . . . 4.2 Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 The Primary Independent Variable: Race/Hispanic Origin . 4.5 The Primary Dependent Variable: Time to All-Cause Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 The Primary Covariate: Year of Birth . . . . . . . . . . . . . . . 4.7 Other Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7.1 Educational Attainment . . . . . . . . . . . . . . . . . . . . 4.7.2 Body Mass Index (BMI) . . . . . . . . . . . . . . . . . . . 4.7.3 Number of Career Years Played in Major League Baseball . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7.4 Player Position . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7.5 US Birthplace Region . . . . . . . . . . . . . . . . . . . . . 4.7.6 Handedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.8 Statistical Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.9 SAS Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

33 33 34 34 34

. . . . .

. . . . .

. . . . .

. . . . .

36 38 38 38 39

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

40 40 40 41 41 43 43

Births and Deaths in the Study Cohort . . . . . . . . . . . . . . . . . . . . 5.1 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.1 Cohort Description . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.2 Year of Birth by Race/Hispanic Origin . . . . . . . . . . 5.1.3 Unadjusted Analysis of Vital Status by Race/Hispanic Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.4 Vital Status by Race/Hispanic Origin Adjusted for Year of Birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.5 Vital Status by Race/Hispanic Origin Stratified by Year of Birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Summary of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

45 45 45 45

..

46

..

47

.. ..

47 48

Part II 4

5

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

Empirical Study Description and Preliminary Results

Contents

6

Preliminary Testing of Main Study Hypotheses . . . . . . . . . . . 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1 Testing of Differences in Mortality Risk Based on Race/Hispanic Origin . . . . . . . . . . . . . . . . . . . . . 6.2.2 Summary of Empirical Results . . . . . . . . . . . . . .

Part III 7

8

xix

.... .... ....

49 49 50

.... ....

50 51

Other Independent Variables and More Definitive Testing of Study Hypotheses

The Role of Educational Attainment . . . . . . . . . . . . . . . . . . . . . . 7.1 Education, Race, and National Origin . . . . . . . . . . . . . . . . . 7.2 Historical Educational Attainment Statistics by Race/Hispanic Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Education and Mortality in the Scientific Literature . . . . . . . . 7.4 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.1 Educational Attainment and Race/Hispanic Origin . . 7.4.2 Educational Attainment and Mortality Risk . . . . . . . 7.4.3 Does Educational Attainment Affect the Relationship Between Race/Hispanic Origin and Mortality Risk? . 7.5 Summary of Empirical Results . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Examination of Other Independent Variables . . . . . . . . . . . . . 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Body Mass Index (BMI) . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.2 BMI and Race/Hispanic Origin in the General Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.3 BMI and Mortality in the General Population . . . 8.2.4 BMI and Professional Athletes . . . . . . . . . . . . . . 8.3 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.1 BMI and Race/Hispanic Origin . . . . . . . . . . . . . . 8.3.2 BMI and Mortality Risk . . . . . . . . . . . . . . . . . . . 8.3.3 Does BMI Affect the Relationship Between Race/ Hispanic Origin and Mortality Risk? . . . . . . . . . . 8.3.4 Summary of Empirical Results . . . . . . . . . . . . . . 8.4 US Birthplace Region . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5.1 US Birthplace Region by Race . . . . . . . . . . . . . . 8.5.2 US Birthplace Region by Mortality Risk . . . . . . .

.. ..

55 55

. . . . .

. . . . .

56 56 58 58 59

.. .. ..

59 60 61

. . . .

. . . .

. . . .

. . . .

63 63 63 63

. . . . . .

. . . . . .

. . . . . .

. . . . . .

64 64 65 66 66 66

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

67 68 68 68 69 69 69

xx

Contents

8.5.3

Does US Birthplace Region Affect the Relationship Between Race and Mortality Risk? . . . . . . . . . . . . . 8.5.4 Summary of Empirical Results . . . . . . . . . . . . . . . . 8.6 Total Number of Years Played in Major League Baseball . . . 8.6.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.6.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7.1 Total Years of Playing Experience and Race/Hispanic Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7.2 Total Years of Playing Experience and Mortality Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7.3 Does Total Years of Playing Experience Affect the Relationship Between Race/Hispanic Origin and Mortality Risk? . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7.4 Summary of Empirical Results . . . . . . . . . . . . . . . . 8.8 Player Position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.8.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.9 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.9.1 Player Position and Race/Hispanic Origin . . . . . . . . 8.9.2 Player Position and Mortality Risk . . . . . . . . . . . . . 8.9.3 Does Player Position Affect the Relationship Between Race/Hispanic Origin and Mortality Risk? . . . . . . . . 8.9.4 Summary of Empirical Results . . . . . . . . . . . . . . . . 8.10 Handedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.10.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.11 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.11.1 Handedness and Race/Hispanic Origin . . . . . . . . . . . 8.11.2 Handedness and Mortality Risk . . . . . . . . . . . . . . . . 8.11.3 Does Handedness Affect the Relationship Between Race/Hispanic Origin and Mortality Risk? . . . . . . . . 8.11.4 Summary of Empirical Results . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Final Multivariate Testing of Study Hypotheses . . . . . . . . . . . . 9.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.1 Multivariate Analysis Using Categorical Variables . 9.2.2 Multivariate Analysis Using Continuous Variables . 9.3 Post hoc Analysis of Mortality Risk Between Hispanic and Non-Hispanic White Players Stratified by Education . . . . . . 9.3.1 Summary of Empirical Results . . . . . . . . . . . . . . .

. . . . . .

70 70 71 71 72 72

..

72

..

73

. . . . . . .

. . . . . . .

73 74 74 74 76 76 77

. . . . . . .

. . . . . . .

77 78 79 79 80 80 80

.. .. ..

81 81 81

. . . . .

. . . . .

85 85 85 85 86

... ...

88 89

. . . . .

. . . . . .

Contents

Part IV

xxi

External Analysis and Discussion

10 Comparison of Mortality Rates Between Major League Baseball Players and the General Population . . . . . . . . . . . . . . . . . . . . . . 10.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.1 African-American Players Versus African-Americans in the General Population . . . . . . . . . . . . . . . . . . . . 10.3.2 Hispanic Players Versus Hispanics in the General Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.3 Non-Hispanic White Players Versus Whites in the General Population . . . . . . . . . . . . . . . . . . . . . . . . . 10.4 Empirical Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Summary, Conclusions, and Implications . . . . . . . . . . . . . . 11.1 Internal Comparison Summary and Implications . . . . . . 11.1.1 African-American Versus Non-Hispanic White Mortality Comparisons . . . . . . . . . . . . . . . . . . 11.1.2 Hispanic Versus African-American Mortality Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . 11.1.3 Non-Hispanic White Versus Hispanic Player Mortality Comparisons . . . . . . . . . . . . . . . . . . 11.2 The Hispanic Paradox . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 Other Independent Risk Factors of Mortality . . . . . . . . 11.3.1 Educational Attainment . . . . . . . . . . . . . . . . . . 11.3.2 BMI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.3 Pitching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.4 Years of Playing Experience . . . . . . . . . . . . . . 11.4 Other Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4.1 US Birthplace Region . . . . . . . . . . . . . . . . . . . 11.5 Summary and Implications of External Results . . . . . . . 11.5.1 Healthy Worker Effect . . . . . . . . . . . . . . . . . . 11.5.2 The US General Population: A Low Bar for Comparisons with Professional Athletes . . . . . . 11.6 Study Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.7 Final Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

93 93 94 95

..

97

..

97

.. .. ..

97 98 98

. . . .

. . . . . . 101 . . . . . . 101 . . . . . . 101 . . . . . . 104 . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

104 105 106 106 107 108 109 110 110 111 111

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

112 112 113 114

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

List of Figures

Fig. 4.1

Census regions and divisions of the United States . . . . . . . . . . . .

41

xxiii

List of Tables

Table 3.1 Table 3.2

Table 5.1 Table 5.2 Table 5.3 Table 6.1

Table 7.1

Table 7.2 Table 7.3

Table 7.4

Table 8.1 Table 8.2 Table 8.3

Life expectancy at birth (in years) for US males between 1970 and 2014 by race/Hispanic origin . . . . . . . . . . . . . . . . Age-adjusted all-cause, heart disease and malignant neoplasm death rates per 100,000 US resident population for selected years between 1970 and 2014 by race/Hispanic origin (males) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Year of birth by race/Hispanic origin . . . . . . . . . . . . . . . . . . Vital status by race/Hispanic origin . . . . . . . . . . . . . . . . . . . Vital status by year of birth categories and race/Hispanic origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazards regressions predicting mortality risk by race/Hispanic origin controlling for year of birth with rotating index reference groups . . . . . . . . . . . . . . . . . . Median number of school years completed in US by race and Hispanic cohort countries, 1940–1970 (males aged 25 years and older) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Educational attainment of Major League Baseball players by race/Hispanic origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on educational attainment controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on educational attainment and race/Hispanic origin controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . . Categorical, mean and median BMI by race/Hispanic origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazard regression predicting mortality risk based on BMI controlling for year of birth . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on (categorical) BMI and race/Hispanic origin controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . .

..

23

.. .. ..

24 46 47

..

48

..

50

..

57

..

58

..

59

..

60

..

66

..

67

..

67 xxv

xxvi

Table 8.4 Table 8.5

Table 8.6

Table 8.7 Table 8.8

Table 8.9

Table 8.10 Table 8.11 Table 8.12

Table 8.13 Table 8.14 Table 9.1

Table 9.2

Table 10.1

Table 11.1

List of Tables

US birthplace region by race . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on US birthplace region controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on US birthplace region and race controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total number of Major League Baseball years played by race/Hispanic origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on total number of Major League Baseball years played controlling for year of birth . . . . . . . . . . . . . . . . . . . Cox proportional hazards regression predicting mortality risk based on total number of Major League Baseball years played and race/Hispanic origin controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Player position by race/Hispanic origin . . . . . . . . . . . . . . . . Cox proportional hazards regressions predicting mortality risk based on player position controlling for year of birth . . Cox proportional hazards regressions predicting mortality risk based on player position and race/Hispanic origin controlling for year of birth . . . . . . . . . . . . . . . . . . . . . . . . . Handedness by race/Hispanic origin . . . . . . . . . . . . . . . . . . . Cox proportional hazards regressions predicting mortality risk based on handedness controlling for year of birth . . . . . Stepwise backwards elimination Cox proportional hazards regression predicting mortality risk using categorical variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stepwise backwards elimination multivariate Cox proportional hazards regression predicting mortality risk using continuous variables . . . . . . . . . . . . . . . . . . . . . . . . . . Standardized mortality rate (SMR) results for comparisons between Major League Baseball players in 3 race/Hispanic origin groups versus their general population counterparts . . Summary of key hazard ratio results for race/Hispanic origin predicting mortality risk with rotating index reference groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

69

..

70

..

71

..

72

..

73

.. ..

74 76

..

77

.. ..

78 80

..

81

..

87

..

88

..

96

. . 102

Part I

The Backstory

Chapter 1

The Roots of Baseball in the US

1.1 Baseball in the US Dating the origins of baseball depends largely on how one defines “baseball” (Vaught 2013). Ball and bat games can be traced back several millenniums to possible origins in Egypt around 1500 B.C. Fast forwarding 2500 years, ball and bat games were played in England following the Norman Invasion in 1066 (Vaught 2013). Remotely resembling baseball, a combination of 2 English games, rounders and cricket, was observed in the US during the eighteenth century (History Staff 2013). However, these early games had few similarities to baseball as it began to be played in the mid-nineteenth century. Myths about the origins of modern baseball in the US are not uncommon. Baseball historians unanimously agree that the game was not invented by Abner Doubleday in Cooperstown, New York in 1839, as some business officials wanted fans to believe. In fact, Doubleday, a decorated Civil War Union Army General, was proclaimed the inventor of baseball as a nationalistic ploy at a meeting of Spalding marketing executives held in 1907 (McDonald 2016). The truth is that Doubleday never had anything to do with the game of baseball (History Staff 2013). The patriotic facade of baseball was based on an American longing to have a sport that was native-born and raised—not imported from Europe or elsewhere. Thought by many to be a democratizing game, baseball would become, and still is, the national pastime in the US. In 1889, Mark Twain described baseball as “… the very symbol, the outward and visible expression of the drive, the push, the rush and struggle of the raging, tearing, booming nineteenth century” (Twain 1889). During the early and mid-nineteenth century, baseball in the US was a social activity played by men as a form of leisure. In those days, fielders would get a runner out by “plugging” or “soaking” him, which meant throwing and hitting baserunners with the hard ball. The new so-called Knickerbocker Rules (Thorn 2004) did away with plugging which in turn helped reduced arguing and violence on the playing field (Baseball Reference 2018). In addition to tagging baserunners, or having the © The Author(s), under exclusive license to Springer Nature Switzerland AG 2019 J. S. Markowitz, Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball, SpringerBriefs in Public Health, https://doi.org/10.1007/978-3-030-17280-0_1

3

4

1 The Roots of Baseball in the US

ball beat them to a base, the Knickerbocker Rules contained additional elements of baseball as played today, such as 90 feet between bases within a diamond-shaped infield, playing teams comprised of 9 players, foul lines, a 3-strike rule, and games that were generally 9 innings in length. The first baseball team in the US in 1845, the New York Knickerbockers, was comprised of amateurs from an upper-class New York City Social Club (Thorn no date). Their first game in 1846, was played against a cricket team (History Staff 2013). Other baseball clubs emerged during this period, often along racial and ethnic lines (Ashe Jr. 1988). Initially, baseball was played mainly in the Northeastern US, although Civil War soldiers, who played the game in-between battles, helped spread the game throughout the country (Ruck 2011). By 1857, there were 16 New York City area teams that formed the first amateur baseball organization, the National Association of Base Ball Players, or NABBP (Wright 2000). As the governing body of baseball, the NABBP organized the first championship games. By 1865, there were about 100 NABBP baseball teams and within the next 2 years, the number of teams quadrupled. While some star players were secretly paid for playing the game in the early days of the sport, baseball teams were primarily comprised of individuals who did not receive any compensation. This changed in 1869 when 12 teams from the NABBP announced that they would turn professional. In 1871, the NABBP was replaced by a professional baseball league known as the National Association of Professional Base Ball Players, often called the National Association. In 1876, the National Association was replaced by the National League. The American League of Professional Baseball Clubs, later known as the American League, was formed in 1901 as the National League’s rival. Two years later, the winners of the National and American Leagues competed in the first World Series. While many changes to the organization and structure of Major League Baseball have occurred since the beginning of the twentieth century, these are the same National and American Leagues that compete today.

1.2 African-Americans and Slavery in the US The well-established issues that provoked the Civil War included states’ rights, trade, tariffs, and especially slavery. Because baseball in the US took root before, during and after the Civil War, a brief account of slavery in the US is relevant. In 1619, the first captive Africans were brought to Jamestown, Virginia by the Dutch and the degrading practices of slavery became prevalent in all US colonies. As a cheap and abundant source of labor, slaves worked mainly on large plantations or small farms, producing tobacco, indigo, rice, and cotton (after the invention of the cotton gin in 1793). By the time the African slave trade was outlawed in 1808, more than 400,000 slaves had been brought to the US. Despite this, a domestic slave trade persisted, resulting in a dramatic increase in the number of slaves. By 1860, it’s estimated that about 4 million slaves resided in the US with about half in Southern states that produced cotton (History.com a no date).

1.2 African-Americans and Slavery in the US

5

Because of poor nutrition, impoverished living conditions, and dangerous work, slaves suffered disproportionately from a variety of conditions and diseases (Downs 2012). Sometimes referred to as “Negro diseases,” these conditions included tetanus, cholera, tuberculosis, influenza, hepatitis and syphilis (Kiple and King 1981). When slaves became sick, their treatment was minimal, if not altogether absent. Not unexpectedly, slaves had short life expectancies. Only 4 of every 100 slaves in the US lived to 60 years of age or older (Frederick Douglass Family Initiatives 2018).

1.3 African-Americans in Baseball Prior to the turn of the twentieth century, there were isolated periods when whites and African-Americans played organized baseball together. However, up until the time of Jackie Robinson’s appearance in 1947, baseball in the US was a largely segregated game. With few exceptions, integrated baseball was only played in the more tolerant sections of the US, i.e., the Northeast and Midwest. The end of the Civil War and Reconstruction generated monumental hardships for African-Americans, particularly in the South. Nevertheless, organized amateur baseball became popular in the immediate aftermath of the war. During the next 20 years, roughly 200 allAfrican-American teams were organized throughout the country (PBS 2014). According to the Baseball Reference history section: “As early as 1867, the racism of the post-Civil War era showed up in the national pastime: The National Association of Baseball Players … voted to exclude any club that had black players from playing with them” (Baseball Reference 2018). This period also saw the rise of Jim Crow laws in the US (especially in the South), which segregated the 2 races in most aspects of living and “… forbade whites and Blacks [from] competing in the same athletic contests” (Tygiel 1983). By 1872, with the disbandment of the NABBP, several African-American players appeared on the heretofore all white teams and became the first blacks in professional baseball. Several African-Americans played professional baseball together with whites during the 1870s and 1880s. However, [African-American players] “… were shunned by teammates, vilified by spectators, and brutalized by opponents” (Lamb 2012). It was not uncommon for African-Americans to be targets of physically abusive play by white players during the games they played together. For example, white pitchers would sometimes intentionally throw balls at African-American batters’ heads (McKissack and McKissack 1994; White 1907), a practice known as “bean-balling.” The intentional spiking of African-American infielders by white baserunners also became a frequent practice in the integrated games. It was an act of self-protection when 2 of the first African-American professional baseball players, Bud Fowler and Frank Grant, both infielders, invented shin guards to help minimize injury when intentionally spiked by white players sliding into bases (Peterson 1970). In the late nineteenth century, some white players engaged in practices to remove African- Americans from professional baseball games. For example, at an exhibition game in 1883, Cap Anson, a white player/manager for the Chicago White Stockings,

6

1 The Roots of Baseball in the US

and a future Hall of Famer, refused to take the field if the African-American catcher for the Toledo team was permitted to play. Again in 1887, Anson demanded that 2 African-American players on the opposing Newark team not be allowed to play in a scheduled game. Almost immediately after Anson’s refusal to play against AfricanAmerican players in 1887, directors of the International League, a vital Minor League baseball organization, “… decided that since white players had misgivings about blacks and had threatened to leave the league if something was not done, no more contracts would be offered to black ball players” (McKissack and McKissack 1994). With no contract signings of African-Americans in baseball’s Minor League system, blacks were phased out of Major League Baseball. By the end of the 1880s only a few African-Americans were playing Minor League baseball (Tygiel 1983). While never officially written, Major League Baseball’s color line was codified by a policy known as the “Gentleman’s Agreement” which effectively banned AfricanAmericans from the game. Even the Commissioner of baseball, a former US federal judge named Kenesaw Mountain Landis, never acknowledged an official ban of African-Americans players in the League (Gurevitz 2015). Given the subterfuge in establishing Major League Baseball’s color line, its start date is thought to be around 1892. Although slavery had been abolished in the US and Constitutional amendments passed promising equal rights, African-Americans were barred from the country’s national sport. Only with the introduction of Jackie Robinson into Major League Baseball in 1947 would the color barrier come to an end.

1.4 Baseball and the Larger American Society During the years immediately following the Civil War there was a glimmer of hope “… that baseball—like America—could achieve its promise of equality for all” (Burgos Jr. 2007). After all, slavery had been abolished and amendments to the Constitution had provided African-Americans with citizenship, equal protection, due process, as well as the right to vote. Unfortunately, this legitimacy turned out to be more theoretical than real, particularly in the Southern part of the US, due to laws and policies that were intimidating and segregating. Even the US Supreme Court played a leading role in maintaining separation between African-Americans and whites. The infamous Plessy v. Ferguson Supreme Court ruling of 1896 legalized the doctrine of “separate but equal” (Legal Information Institute no date). While the “separate” part of this ruling was extolled by many whites, the “equal” part was either ignored or minimized by other rules and policies that ensured inequality for African-Americans. After their forced removal from the Major Leagues, African-American professional players participated on predominantly black teams in organized baseball games. Perhaps the most famous of these teams, the Cuban Giants, was initially formed in 1885. Remarkably, several of the all- or predominantly- African-American teams were named “Cubans.” This was intended to fool some opponents and fans

1.4 Baseball and the Larger American Society

7

into thinking that the players were Cuban rather than African-American. It has even been reported that “… [Cuban Giants] players chatted [Spanish] gibberish on the field to pass as Cubans” (Tygiel 1983). African-American baseball teams during this period were plagued by unfair financial agreements. “Gate receipts were tightly controlled by white booking agents. The agents dictated when and where black teams could play, and they subsequently passed little of the games’ attendance revenues on to team owners” (Kelly no date). In part, the frustration that arose from the unfair financial treatment of African-American owners and players, motivated Rube Foster, the “father of Black baseball” and a future Hall of Famer, to form the Negro Major League in Kansas City, Missouri in 1920. Branching out from the Midwest, at least 7 Negro Leagues emerged in the East and the South in both urban and rural areas. Negro League baseball teams often participated in “barnstorming” games, traveling throughout the country and Latin America to play non-league games for pay. Negro League teams often played day games and traveled at night. As the financial pressures of Negro League players and owners intensified, some teams played both day and night games on the same day, even in different cities. Because Negro League teams rarely owned their own stadiums, they rented venues from Major and Minor League teams. This money often represented a significant percentage of income for these organizations. For example, the New York Yankees alone made about $100,000 per year based on revenues from Negro League baseball (Bradbury 2007). Despite this, African-American baseball players were often not permitted to use the regular (all-white) locker rooms and showers. In addition, Negro League games were sometimes cancelled at the last minute by the Major and Minor League organizations when they were offered more lucrative situations. By the 1930s, some of the greatest baseball players of all-time played in Negro Leagues. Nevertheless, it is generally accepted that the overall quality of Major League Baseball play was better than the Negro Leagues. Several African-Americans superstar players, like Satchel Paige and Josh Gibson, earned salaries that eclipsed those of most white baseball players. On average, however, the salaries earned by Major League Baseball players, exceeded those earned by Negro League players. According to one author: “… most [African-American] players earned between $125 and $300 a month, less than half the average white athlete” (Tygiel 1983). Many African-American players also played baseball during the winter months in professional leagues in Latin American countries. Notably, average salaries for Negro League baseball players exceeded those earned by African-Americans in the general population (Tygiel 1983).

1.5 The Role of World War II Some writers and historians viewed the oppression of Jews in Europe as partly analogous to the situation experienced by African-Americans in the US. In many ways, American involvement in World War II was paradoxical: while the US was

8

1 The Roots of Baseball in the US

fighting for freedom and equality abroad, many African-Americans were denied these same basic rights at home. Just several weeks after the Japanese attack on Pearl Harbor, the Pittsburgh Courier launched the “Double V” campaign—victory in the war being fought against the Axis powers, plus victory at home in the fight against racism and inequality. The Double V campaign encouraged African-Americans to serve in the US Armed Services during World War II and simultaneously called on them to demand implementation of the 3 post-Civil War equal rights amendments. The Double V campaign sought to fight fascism in Europe as well as racism, segregation and unequal rights in the US. As in the broader US society, Jim Crow laws were pervasive in the military during the war. Although African Americans could volunteer for military service and even be drafted, when permitted to serve, they were relegated to segregated units and usually given menial jobs. “The world’s greatest democracy fought the world’s greatest racist with a segregated army” (Ambrose 1997). Prior to World War II, less than 4000 African-Americans served in the US Armed Forces and there were hardly any African-American officers. By the end of the war in 1945, over a million African-Americans had served in the military with many integral to fighting and winning the war. Several segregated all- African-American military units were nationally acclaimed. For example, the Tuskegee Airmen flew more than 15,000 sorties in North Africa, Sicily and Europe and distinguished themselves as courageous, world-class airmen (Manning 2012). World War II had laid the groundwork for ending segregation in Major League Baseball.

1.6 Desegregation in Major League Baseball Many social, political and economic forces impacted the battle over AfricanAmericans’ re-entry into Major League Baseball, and by the 1930s, the pressures to desegregate had intensified. The movement to integrate baseball was led by a variety of individuals and groups that included African-American and progressive journalists, white activists and civil rights groups. The movement occurred around the same time that other anti-discrimination struggles were underway related to housing, jobs, and education (Dreier 2013). With the extensive involvement of African-Americans in World War II, tensions related to desegregating baseball increased in the 1940s. Clearly, many AfricanAmerican baseball players were talented enough to play in the Major Leagues and many of them were already stars in Negro Leagues. Ostensibly, talented AfricanAmerican players would improve the chances of Major League teams to win games, and perhaps even championships. Of most importance for some, ending segregation in Major League Baseball was the “morally right thing to do.” While the separate but equal doctrine created by the Plessy v. Ferguson case dominated US race culture and policy, significant exceptions in the sports world had already become evident. For example, African-Americans represented the US at the 1932 and 1936 Olympic

1.6 Desegregation in Major League Baseball

9

games and successfully participated in numerous college sports, like baseball and football. Yet, some Major and Minor League baseball team owners feared that the Negro Leagues would not survive if African-Americans were permitted to play Major League Baseball. Unsubstantiated warnings were circulated around baseball that racial incidents would occur if African-Americans were permitted to play in the Majors. Moreover, there were more than a few Major League Baseball team owners, other executives and players who had anti-African-American feelings and beliefs. For example, one team owner, Tom Yawkey, grew up on a large plantation in South Carolina and strongly opposed the idea of ending segregation in Major League Baseball. His team, the Boston Red Sox, became the last to sign an African-American to a Major League Baseball contract (Kepner 2017), 12 years after Jackie Robinson broke the color line.

1.7 Jackie Robinson and the Desegregation of Baseball Jackie Robinson, the grandson of a slave, was born on January 31, 1919 in a plantation shack in Cairo, Georgia. Jackie, the youngest of 5 children, was only a year old when his father left the family. Robinson’s mother moved the family to Pasadena, California, where she worked a variety of jobs to sustain her family. As a child and young man in both Cairo and Pasadena, Robinson and his family faced numerous incidents involving segregation and discrimination (Tygiel 1983). Robinson excelled at multiple sports in high school and college including football, baseball, tennis and track. Prior to being drafted into a segregated US Army unit in 1942, he played semi-pro football. Robinson went to Officer Training School, a rare feat for an African-American at the time, and in 1943, he became a 2nd Lieutenant in the US Army. However, Robinson’s military career ended abruptly the next year following a racial incident when he was ordered, but refused, to move to the back of an Army bus. Court martial proceedings ensued on charges of “insubordination during questioning.” Despite his acquittal, the Army would not allow Robinson to deploy overseas. Several months later, Robinson was honorably discharged. In 1945, Robinson was offered a contract to play baseball professionally in the Negro League. Following 5 months of play for the Kansas City Monarchs, Robinson was signed to a contact with the Montreal Royals of the Brooklyn Dodgers’ Minor League organization. The primary engineer of Robinson’s ascent to Major League Baseball was Dodgers’ team President and General Manager, Branch Rickey, a former college baseball coach. It is believed that Rickey was motivated to integrate baseball both morally, based on incidents he experienced while coaching, as well as economically. As part of the Great Migration, beginning around 1916 and continuing through 1970, about 6 million African-Americans transplanted themselves from Jim Crow dominated regions in the rural South to Northern, Midwestern and Western cities (Wilkerson 2016). As African-Americans populated large cities outside the South, many were known to watch and enjoy Negro League baseball. Thus, Rickey

10

1 The Roots of Baseball in the US

believed that signing Robinson would increase Dodger ticket sales. Rickey also felt that Robinson’s participation would improve the team and perhaps help the Dodgers win a championship, a feat that had eluded the team since their inception in 1883. During his 10-year career with the Dodgers, Robinson led the team to 6 National League pennants as well as their first World Series championship title in 1955. Inducted into the Baseball Hall of Fame during his first year of eligibility in 1962, the League also retired his jersey number 42 from all Major League Baseball, the only number to be honored in this manner. [Robinson’s] “… dignified courage in the face of virulent racism—from jeers and insults to beanballs, hate mail, and death threats—commanded the admiration of whites as well as blacks and foreshadowed the tactics that the 1960s Civil Rights Movement would develop into the theory and practice of nonviolence” (History.com b no date). Jackie Robinson died in October 1972 at the age of 53 from heart disease and diabetes complications. The genesis and early development of professional baseball in the US coincided with massive political and cultural changes that fostered between-race inequalities that would persist for decades. Legal attempts to resolve these matters often met with resistance that made progress slow and difficult. African-American service during World War II was part of a social and political movement that helped set the stage for desegregating Major League Baseball. Rather than ending between-race inequalities, including health and mortality disparities, desegregation in baseball set in motion new changes and challenges. Hispanics also entered the US professional baseball fray in earnest beginning in 1947 with an influx of players that continues robustly today. A brief account of Latin American history, particularly as it relates to race and social class, follows in Chap. 2.

References Ambrose, S. E. (1997). Citizen soldier. The US Army from the Normandy beaches to the bulge to the surrender of Germany. New York: Simon & Schuster. Ashe, A., Jr. (1988). A hard road to glory: Baseball. New York: Amistad Press Inc. Baseball Reference (2018): History of baseball in the United States. Retrieved from https://www. baseball-reference.com/bullpen/History_of_baseball_in_the_United_States. Burgos, A., Jr. (2007). Playing America’s game: Baseball, Latinos, and the color line. Berkeley: University of California Press. Bradbury, J. C. (2007). The baseball economist: The real game exposed. New York: Penguin Group. Downs, J. (2012). Sick from freedom: African-American illness and suffering during the Civil War and Reconstruction. New York: Oxford University Press. Dreier, P. (2013). The real story of baseball’s integration that you won’t see in 42. The Atlantic. Retrieved from https://www.theatlantic.com/entertainment/archive/2013/04/the-realstory-of-baseballs-integration-that-you-wont-see-in-i-42-i/274886/. Frederick Douglass Family Initiatives. (2018). Additional facts provided by the Gilder Lehrman Institute of American History. Retrieved from http://www.fdfi.org/additional-facts.html. Gurevitz, A. E. (2015). Breaking Baseball’s color line. Arizona Journal of Interdisciplinary Studies, 4, 86–101.

References

11

History.com a. (no date). Slavery in America. Retrieved from https://www.history.com/topics/blackhistory/slavery. History.com b (no date). Jackie Robinson. Retrieved from https://www.history.com/topics/blackhistory/jackie-robinson. History Staff. (2013). Who invented baseball? Retrieved from https://www.history.com/news/askhistory/who-invented-baseball. Kelly, M. (no date). The Negro National League is founded. National Baseball Hall of Fame. Retrieved from https://baseballhall.org/discover-more/stories/inside-pitch/negro-nationalleague-is-founded. Kepner, T. (2017). Thomas Yawkey, the Red Sox owner who resisted integration. New York Times. Retrieved from https://www.nytimes.com/2017/08/18/sports/baseball/red-sox-thomas-yawkeyintegration.html?_r=0. Kiple, K. F., & King, V. H. (1981). Another dimension to the black diaspora. UK: Cambridge University Press. Lamb, C. (2012). Conspiracy of silence: Sportswriters and the long campaign to desegregate baseball. Lincoln: University of Nebraska Press. Legal Information Institute. (no date). Plessy v. Ferguson. Retrieved from https://www.law.cornell. edu/supremecourt/text/163/537. Manning, R. (2012). The ‘Tuskegee Experiment’: The legend of the Tuskegee Airmen. Retrieved from http://www.malmstrom.af.mil/News/Features/Display/Article/349719/thetuskegee-experiment-the-legend-of-the-tuskegee-airmen/. McKissack, P. C., & McKissack, F., Jr. (1994). Black diamond: The story of the Negro baseball Leagues. New York: Scholastic Inc. McDonald, G. (2016). Who really invented baseball? Retrieved from https://www.seeker.com/whoreally-invented-baseball-1771280977.html. PBS. (2014). African-American baseball. Retrieved from http://www.pbs.org/opb/ historydetectives/feature/african-american-baseball/. Peterson, R. (1970). Only the ball was white: A history of legendary black players and all-black professional teams. Oxford: Oxford University Press. Ruck, R. (2011). Raceball: How the Major Leagues colonized the black and Latin game. Boston: Beacon Press. Thorn, J. (no date). October 1845: The first recorded baseball games in New York. Retrieved from https://sabr.org/gamesproj/game/october-1845-first-recorded-baseball-games-new-york. Thorn, J. (2004). Total baseball: The ultimate baseball encyclopedia. Wilmington: Sport Media Pub. Twain, M. (1889). TwainQuotes.com. Retrieved from http://www.twainquotes.com/Baseball.html. Tygiel, J. (1983). Baseball’s great experiment: Jackie Robinson and his legacy. Oxford: Oxford University Press. Vaught, D. (2013). The Farmers’ game: Baseball in rural America. Baltimore: Johns Hopkins University Press. White, S. (1907). Sol White’s history of colored baseball with other documents on the early black game, 1886–1936. Omaha: University of Nebraska Press. Wilkerson, I. (2016). The long-lasting legacy of the Great Migration. Retrieved from https://www. smithsonianmag.com/history/long-lasting-legacy-great-migration-180960118/. Wright, M. D. (2000). The National Association of Base Ball Players, 1857–1870. Jefferson: McFarland & Company.

Chapter 2

A Mini-history of Latin America and Professional Baseball

2.1 Background It is beyond the scope of this book to detail the history of all countries where Latin American Major League Baseball players in the current study cohort were born. Briefly, these countries included Cuba, the Dominican Republic, Venezuela, Puerto Rico, Mexico and Panama. (Puerto Rico is a US territory, although for study purposes, it will be labeled a “country.”) In addition, a very limited number of players in the study cohort were born in Nicaragua and Colombia. These 8 countries will be referred to in this book as the “Hispanic cohort countries.” The countries of Latin America are diverse, and the cohort countries reflect this. Nevertheless, the cohort countries share common historical threads. As was the case in the US, all the Hispanic cohort countries were conquered and colonized by European powers beginning in the fifteenth century. While the US was colonized by the British, the Hispanic cohort countries were colonized by the Spaniards. The conquered territories were initially inhabited by indigenous populations who had virtually no previous contact with Europeans. Lacking immunity for common diseases experienced in Europe, up to 90% of indigenous populations in Latin America died due to epidemics of smallpox, typhus, influenza, measles and several other diseases (Montenegro and Stephens 2006). To meet cheap laborer needs resulting from widespread deaths among indigenous people, the European colonizers began importing slaves from Africa. The production of crops, such as sugar, tobacco, coffee and cotton, mainly on plantations that utilized slave labor, became the foundation of many Latin American economies. The Atlantic slave trade continued for more than 350 years (Gilder Lehrman Institute of American History no date). While slavery had developed in Africa prior to the start of the slave trade, it was more like European serfdom and not nearly as malevolent as the form of slavery that subsequently emerged in the US (Davidson 1988). Moreover, while Africans living in Spanish colonized countries were considered property, they were sometimes charged with responsibilities that enabled them to function as intermediaries © The Author(s), under exclusive license to Springer Nature Switzerland AG 2019 J. S. Markowitz, Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball, SpringerBriefs in Public Health, https://doi.org/10.1007/978-3-030-17280-0_2

13

14

2 A Mini-history of Latin America and Professional Baseball

between Europeans and indigenous people. A limited number of African slaves in Latin American countries were able to buy their way out of bondage (Pino 1997). Trading African slaves and their use as laborers proved to be an economically profitable business and by 1870, 7 mainly European countries had shipped about 12.5 million Africans to North America, Latin America and the Caribbean, although about 2 million of these individuals died in route to their targeted locations (Gates Jr. 2014). By the late seventeenth and early eighteenth centuries, pressure mounted on Spain and other European countries to end the slave trade, and eventually, to free the slaves. With strong abolitionist and diplomatic efforts, as well as the emergence of slave revolts throughout the New World, the legal end of slavery came to most Latin American countries in the mid-1800s. The US in 1865, Cuba in 1886, and Brazil in 1888 were the last countries in the New World to abolish slavery (Encyclopedia.com 2006). Moreover, as was the case in the US, the legal abolition of slavery did not result in the end of racism, discrimination and inequalities that disadvantaged Afro-Latinos (Andrews 2018). Social and sexual relations between Europeans, indigenous people, and Africans in Latin America were common and resulted in racially and ethnically blended offspring. The Spanish conquerors as well as the Church sought biological and cultural purity, but the mixing of the different populations made this difficult. In fact, the Spaniards implemented elaborate systems to track the race and ethnicity of Latin American inhabitants by attempting to monitor the genealogy of every person (Martinez 2008). However, in addition to incomplete and inconsistent records, each successive generation resulted in more complicated combinations of intermixed people, and eventually, racial classification systems were abandoned. The legal abolition of slavery in Latin America was deceptive since it failed to provide social, political or economic equality for Afro-descendants, and the racial and class inequalities linked to the slave economies persisted. In territories subjugated by Spain, the conquerors occupied the top realm in a social pyramid that pushed many Blacks and indigenous people to the bottom. The pyramid’s middle level was tied to “…ancestry, appearance… occupation and wealth” (Wade 2008). Compared to those occupying the top and bottom rungs of this class system, placement in the middle was ambiguous and contentious in Latin America.

2.2 Race/Ethnicity and Discrimination in Latin America Versus the US While race in the US developed into a white and African-American dichotomy, Latin America evolved as “a continuum of color” (Eakin 2007). Because race was more complex and nuanced in Latin America relative to the US, there was more fluidity with respect to race and social class. And when race was used as justification to discriminate, it was not always clear who would be affected, particularly among individuals in the middle part of the Latin American stratification system. “If there

2.2 Race/Ethnicity and Discrimination in Latin America Versus the US

15

is little agreement on who is black (or white or indigenous), how can discrimination take place in any systematic way? In contrast, in the US, there was generally a clearer definition of racial identity, based on a few key categories: black, white, Native American …” (Wade 2008). The stratification system that would continue in many Latin American countries following the end of formal slavery placed Afro-Latinos and indigenous people, and to a lesser extent, mixed-race individuals, in economically, socially, and politically disadvantaged situations. The absence of educational opportunities, widespread poverty, and underrepresentation in government did little to help those who found themselves at or near the bottom of the stratification system.

2.3 Hispanics in Baseball The following is a brief description of how baseball evolved in 6 of the Hispanic cohort countries, with a special focus on Cuba, the first Latin American country where baseball developed and grew.

2.3.1 Cuba Sent to the US for a good college education, the Guillo brothers are credited with bringing baseball to their native country, Cuba, in 1864, and within a decade, the sport became enormously popular there. While the game began as a rich man’s sport in the US, baseball had mass appeal in Cuba owing to its simplicity and low cost. Around 1880, an amateur Cuban league was created that eventually became the Cuban Winter League. Not so coincidentally, the first Cuban professional League emerged only 2 years after the National League of Major League Baseball was formed. This professional baseball league operated continuously until the rise of Fidel Castro in the late 1950s. Enthusiasm over the game of baseball in Cuba during the latter part of the nineteenth century coincided with its struggle for independence and modernization. Choosing baseball over bullfighting was analogous to choosing independence over Spanish domination and became part of Cuba’s “rehearsal for democracy” (Perez 1994). The earlier Cuban professional baseball teams excluded blacks, but by the beginning of the twentieth century, professional baseball in Cuba became integrated. Cuban baseball teams, comprised of multiracial players, barnstormed and created their own leagues in the US around the beginning of the twentieth century. One team, the Cuban Stars, played so well that they were granted charter membership in the US Negro National League in 1920. In contrast to dark-skinned Afro-Cubans, several Spanish ancestry Cubans with light skin color played in Minor League organizations

16

2 A Mini-history of Latin America and Professional Baseball

within Major League Baseball as early as the beginning of the twentieth century (Eschevarria no date). In the aftermath of political turmoil in Cuba that resulted in Civil War, the US occupied the country between 1906 and 1909. This greatly increased Cuban exposure to US baseball with both Negro League and Major League Baseball teams scheduling exhibition games on the island. In 1911, the first 2 Cuban players were signed to Major League Baseball contracts. While the lighter-skin signees would pass as whites to many, there were other, more talented dark-skinned Hispanic players who failed to get contracts. The “Gentleman’s Agreement” that banned African-Americans from baseball for the first half of the twentieth century applied to anyone with black skin (Negro League Baseball Museum 2006) including many Afro-Latinos. Negro League players especially enjoyed playing baseball in Cuba. While far from an ideal country when it came to race, Cuba emerged as a model for multiracial, international baseball play at the highest levels. “The [Cuban Baseball League] was an anomaly, the only place in the world where the best professional ballplayers of all nations and colors competed with and against each other” (Ruck 2011).

2.3.2 Venezuela Baseball spread to Venezuela during the last decade of the nineteenth century. As was the case in Cuba, it was a group of young men attending US universities who introduced the game to the country. Initially, players were from high-class social clubs in Caracas, but later the game was more broadly adopted. The sport rapidly gained popularity throughout Venezuela and the country’s first baseball stadium was built within a few years. Interestingly, the first organized baseball game in Venezuela was billed as a “new kind of Chess game” (Baseball Reference 2013). By 1920, there were 30 organized baseball teams who played in 10 different ballparks in the city of Maracaibo alone (Baseball Reference 2013). When oil was discovered in Venezuela in 1922, money flowed into the country attracting white and Negro League players from the US to play baseball. By 1927, professional baseball began in Venezuela and in 1945, a professional baseball Winter League was formed. Since 1939, more than 350 baseball players from Venezuela have played in the Major Leagues.

2.3.3 Mexico An early form of baseball may go back to the mid-nineteenth century in Mexico when the game was introduced by US military personnel involved in the MexicanAmerican War. The game spread throughout the country when US soldiers helped lay tracks for a new train system known as the Monterrey-Tampico railway (Hughson 2018). During the latter part of the 1800s, employees of American companies doing business in Mexico helped spread the game throughout the country. Professional

2.3 Hispanics in Baseball

17

baseball in Mexico was first played in 1925. A triple-A Minor League system (the highest level possible) known as the Mexican League, began in 1967 and continues today. Despite intense competition from soccer, baseball persists as a very popular sport in Mexico and contributes dozens of players to Major League Baseball.

2.3.4 Puerto Rico The influence of Cubans and Americans was integral to bringing baseball to Puerto Rico. Because of the Spanish-American War, there were large numbers of American military personnel stationed in Puerto Rico who helped expose and popularize the game. Following the War, a treaty with Spain, would make Puerto Rico a US possession. By 1918, Negro League teams as well as teams comprised of Cuban All-Stars played exhibition games in Puerto Rico. Winter league baseball in Puerto Rico, which began in 1940, continues today. Since the debut of Hiram Bithorn, a Puerto-Rican-born pitcher with the Chicago Cubs in 1942, more than 260 players from the island have played Major League Baseball. In fact, some of the greatest Major League Baseball players of all-time like Roberto Clemente, Orlando Cepeda, Roberto Alomar, and Iván Rodríguez, all Hall of Famers, were born in Puerto Rico.

2.3.5 Dominican Republic The Dominican Republic is another Latin American country in which Cuba played a pivotal role in introducing baseball. In part, this was due to the Ten Years War (1868–1878), an integral part of Cuba’s fight for independence from Spain. When large numbers of Cubans fled the war to the Dominican Republic, they brought with them their love of baseball. By the early 1920s, a professional league comprised of 4 teams formed the Dominican League. A military coup in 1930 led by Rafael Trujillo resulted in the demise of professional baseball in the Dominican Republic around 1937 (MLB.com/DR 2018). Nevertheless, amateur baseball flourished and by 1951 the Winter League of the Dominican Republic was reestablished. Currently, there are more Major League Baseball players born in the Dominican Republic than from any other Latin American country (Wikipedia 2018).

2.3.6 Panama Baseball was played in Panama during the 1880s when the country was a province of Colombia. Upon procuring its independence in 1903, Panama signed a treaty with the US to build a canal beginning the following year. The work on the canal increased Panamanian exposure to baseball particularly among West Indian workers

18

2 A Mini-history of Latin America and Professional Baseball

who participated in the construction effort. Not until 1945 was the Professional Baseball League of Panama established. Although this organization existed continuously through 1972, it subsequently folded due to economic issues, replaced by a strong amateur baseball presence (Reid 2010).

2.4 Journeys Travelled by Hispanic Players to Major League Baseball The journeys travelled by Latin American players to Major League Baseball vary a great deal but have often been marred by personal difficulty and danger. More than a few individuals, including several Hispanic players in the study cohort have risked their lives to play Major League Baseball (Weeks 2017). Many Latin Americans players have emerged from impoverished conditions with little or no education to make it to baseball’s pinnacle. Once they arrive in the US, they can be challenged by cultural differences like dietary changes and language barriers. Being born in Latin America and playing in baseball’s Major League is a tremendous personal and professional accomplishment that often comes with financial rewards that can have life-changing consequences.

References Andrews, G. R. (2018). Inequality: Race, class, gender. In A. de la Fuente & G. R. Andrews (Eds.), Afro-Latin American studies. UK: Cambridge University Press. Baseball Reference (2013). History of baseball in Venezuela. Retrieved from https://www.baseballreference.com/bullpen/History_of_baseball_in_Venezuela. Davidson, B. (1988). The African slave trade. Boston: Little Brown and Company. Eakin, M. C. (2007). The history of Latin America: Collision of cultures. New York: St. Martin’s Griffin. Encyclopedia.com (2006). Emancipation in Latin America and the Caribbean. Encyclopedia of African-American culture and history. Retrieved from https://www.encyclopedia.com/history/ encyclopedias-almanacs-transcripts-and-maps/emancipation-latin-america-and-caribbean. Eschevarria, R. G. (no date). Latin Americans in Major League Baseball through the first years of the 21st century. Encyclopedia Britannica. Retrieved from https://www.britannica.com/topic/ Latin-Americans-in-Major-League-Baseball-910675. Gates Jr., H. L. (2014). How many slaves landed in the US? Retrieved from https://www.theroot. com/how-many-slaves-landed-in-the-us-1790873989. Gilder Lehrman Institute of American History (no date). Historical context: Facts about the slave trade and slavery. Retrieved from https://www.gilderlehrman.org/content/historical-contextfacts-about-slave-trade-and-slavery. Hughson, C. (2018). Baseball in Mexico. Retrieved from http://mopupduty.com/baseball-inmexico-040618/. Martinez, M. E. (2008). Genealogical fictions. Stanford: Stanford University Press. MLB.com/DR (2018). History. Retrieved from http://mlb.mlb.com/dr/history.jsp.

References

19

Montenegro, R. A., & Stephens, C. (2006). Indigenous health in Latin America and the Caribbean. The Lancet, 367(9525), 1850–1869. Negro League Baseball Museum, eMuseum Glossary (2006) Retrieved from http://coe.k-state.edu/ annex/nlbemuseum/glossary.html. Perez, L. A. (1994). Between baseball and bullfighting: The quest for nationality in Cuba, 1868–1898. Journal of American History, 81(2), 493–517. Pino, J. C. (1997). Teaching the history of race in Latin America. Perspectives on history. Retrieved from https://www.historians.org/publications-and-directories/perspectives-on-history/ october-1997/teaching-the-history-of-race-in-latin-america. Reid, L. M. (2010). Baseball’s roots in Panama. Retrieved from https://thesilverpeopleheritage. wordpress.com/2010/01/31/baseballs-roots-in-panama/. Ruck, R. (2011). Raceball: How the Major Leagues colonized the Black and Latin game. Boston: Beacon Press. Wade, P. (2008). Race in Latin America. In D. Poole (Ed.), A companion to Latin American anthropology, Chapter 9. Oxford: Blackwell Publishing Ltd. Weeks, J. (2017). Latino stars in Major League Baseball. Lanham: Rowman and Littlefield. Wikipedia (2018). List of current Major League Baseball players by nationality. Retrieved from https://en.wikipedia.org/wiki/List_of_current_Major_League_Baseball_players_by_ nationality.

Chapter 3

Literature and Statistical Review of Race/Hispanic Origin and Mortality

3.1 General Population Mortality Statistics Prior chapters have provided context with respect to race/Hispanic origin and the evolution of baseball in the US, as well as the Hispanic cohort countries. Now, the focus turns to the primary topic of this book, mortality. Before moving on to the empirical analyses of mortality and race/Hispanic origin among individuals who played Major League Baseball, it is helpful to review the literature and interpret existing statistics related to mortality among the 3 groups of interest in the current study. This chapter is divided into 2 parts addressing mortality as it relates to race and Hispanic origin. The first part of the review focuses on mortality gathered from general population statistics, while the second part examines the mortality literature among professional athletes. The US general population review is derived from vital statistics data, i.e., death certificates, published by the Centers for Disease Control and Prevention (CDC) of the National Center of Health Statistics (2018), hereafter called the “CDC data.” The importance and significance of death certificates from a public health and public policy perspective cannot be emphasized enough. “At both the state and national level, mortality data compiled from death certificates is used to track disease trends, set public health policies, and allocate health and research funding” (Brooks and Reed 2015). There are, however, limitations associated with the use of death certificates. This includes cause of death and ethnicity classification errors. Cause of death errors do not apply to the current study since all-cause mortality is the focus. Ethnicity misclassification has been dropping over time and is currently estimated to be a problem for about 3% of Hispanics (Arias et al. 2016). According to the CDC: “Race and ethnicity reporting on the death certificate continues to be excellent for the white and black populations [and]… is reasonably good for… Hispanic[s]” (National Center of Health Statistics 2018).

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2019 J. S. Markowitz, Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball, SpringerBriefs in Public Health, https://doi.org/10.1007/978-3-030-17280-0_3

21

22

3 Literature and Statistical Review of Race/Hispanic Origin …

In their reporting of vital statistics, the CDC uses terms like ‘Black or AfricanAmericans” and “Hispanic or Latino.” Hence, these same terms will be used whenever the CDC race/Hispanic origin data is cited in the text or table displays in this chapter. Because the current study pertains to male baseball players, the general population literature review will address men only. Together, these 2 mortality-related literature reviews, i.e., general population and professional athletes, should help frame the key study hypotheses that will be specified in Chap. 4.

3.2 CDC General Population Mortality-Related Statistics The review of CDC general population vital statistics includes life expectancy at birth, all-cause death rates and death rates for the 2 most common causes of death in the US, i.e., heart disease and cancer. To show trends over time, statistics are reported for selected years between 1970 and 2014. Prior to 2006, life expectancy statistics for whites and African-Americans included persons of Hispanic and nonHispanic origin. Beginning in 2006, life expectancy statistics are reported separately for “Whites, not Hispanics,” “Blacks, not Hispanics” as well as “Hispanics.” Death rate data for “Black or African-Americans” includes Hispanics and non-Hispanics. Beginning in 1990, death rate data are reported for “Whites, not Hispanic or Latino.”

3.3 Life Expectancy Life expectancy at birth is an estimate of the number of years that an individual or group can expect to live at the time of their birth. This estimate is based on age-specific death rates that are available for each year of birth. According to the CDC “Differences in life expectancy among various demographic subpopulations, including racial and ethnic groups, may reflect subpopulation differences in a range of factors such as socioeconomic status, access to medical care, and the prevalence of specific risk factors in a particular subpopulation” (National Center for Health Statistics 2016). See Table 3.1 for a summary of US general population life expectancy statistics for males for selected years between 1970 and 2014 (National Center for Health Statistics 2016). Life expectancy at birth in 1970 is 68 years for male whites and 60 years for male “Blacks or African-Americans.” By 1980, life expectancy for whites increases to 70.7 years and continues to increase through 2000 to 74.7 years. Life expectancy for Blacks or African-Americans also increases over time to 68.2 years in the year 2000. The between-race gap in life expectancy favoring whites by about 6–8 years fails to change much between 1970 and 2000. Beginning in 2006, the between-race

3.3 Life Expectancy

23

Table 3.1 Life expectancy at birth (in years) for US males between 1970 and 2014 by race/Hispanic origin Race/Hispanic Year origin 1970

1980

1990

2000

2006

2010

2014

Life expectancy Whitea

68.0

70.7

72.7

74.7

See row below

White, not Hispanic

Not available

Not available

Not available

Not available

75.7

Black or AfricanAmericana

60.0

63.8

64.5

68.2

See row below

Black, not Hispanic

Not available

Not available

Not available

Not available

69.5

71.4

72

Hispanic

Not available

Not available

Not available

Not available

77.5

78.5

79.2

76.4

76.5

Source National Center for Health Statistics Health, United States, 2017, Hyattsville, Maryland, 2018, US Government Printing Office, Washington, D.C., Table 15. Located at https://www.cdc. gov/nchs/data/hus/hus15.pdf a Includes persons of Hispanic and non-Hispanic origin

mortality difference decline to 6.2 years, and then further narrow in 2010 to 5 years. By 2014, the between-race difference among US males is at its lowest recorded point of 4.5 years (National Center for Health Statistics 2016). The first publication of life expectancy statistics for Hispanics by the CDC was in 2006. Between 2006 and 2014, life expectancy for Hispanics, relative to non-Hispanic whites, is about 2–3 years older and, relative to Blacks or African-Americans, about 7 years older.

3.4 All-Cause Death Rates Table 3.2 summarizes age-adjusted, all-cause, heart disease and malignant neoplasm mortality rates per 100,000 US general population for male residents for selected years between 1970 and 2014 based on race and Hispanic origin. In 1970, Black or African-American males have an all-cause death rate of 1873.9 per 100,000 residents which is higher than the rates for whites who average 1513.7. In 1990, all-cause death rate data for Hispanics or Latinos are reported for the first time and their rate of 886.4 is far lower than Whites, not Hispanic or Latinos (1170.9) and Black or AfricanAmericans who average 1644.5. By 2014, all-cause death rates drop substantially for all 3 groups with rates of 626.8 for Hispanics or Latinos, 872.3 for Whites, not Hispanic or Latinos and 1034 for Blacks or African-Americans.

24

3 Literature and Statistical Review of Race/Hispanic Origin …

Table 3.2 Age-adjusted all-cause, heart disease and malignant neoplasm death rates per 100,000 US resident population for selected years between 1970 and 2014 by race/Hispanic origin (males) Race/Hispanic origin

Year 1970

1980

1990

2000

2014

Whitea

1513.7

1317.6

1165.9

1029.4

853.4

White, not Hispanic or Latino

Not available

Not available

1170.9

1317.6

853.4

Black or African-Americana

1873.9

1697.8

1644.5

1403.5

1034.0

Hispanic or Latino

Not available

Not available

886.4

818.1

626.8

Whitea

640.2

539.6

409.2

316.7

210.0

White, not Hispanic or Latino

Not available

Not available

413.6

319.9

215.2

Black or African-Americana

607.3

561.4

485.4

392.5

259.5

Hispanic or Latino

Not available

Not available

270.0

238.2

145.7

All-cause death rates

Heart disease death rates

Malignant neoplasms death rates Whitea

244.8

265.1

272.2

243.9

193.0

White, not Hispanic or Latino

Not available

Not available

276.7

247.7

197.7

Black or African-Americana

227.6

291.9

397.9

340.3

231.9

Hispanic or Latino

Not available

Not available

174.7

171.7

135.9

Source National Center for Health Statistics Health, United States, 2017, Hyattsville, Maryland, 2018, US Government Printing Office, Washington, D.C., Tables 21, 22 and 24 Located at https:// www.cdc.gov/nchs/hus/contents2015.htm#017 a Includes persons of Hispanic and non-Hispanic origin

3.5 Heart Disease and Cancer Death Rates Heart disease and cancer consistently represent the top 2 leading causes of death in the US general population. In 2015, about 46–47% of all male deaths in the US are attributable to 1 of these 2 causes with just slight differences in the rates between the 2 diseases. No other cause of death among males in the US general population encompasses more than 7% of all deaths in 2015 (CDC 2018). Male death rate data for heart disease and cancer, the latter referred to as “malignant neoplasms” by the CDC, for selected years between 1970 and 2014 broken down by race and Hispanic origin are displayed in Table 3.2. The trends reflected in this table are very similar to the ones presented previously for overall death rates. With few exceptions, heart disease and malignant neoplasm death rates for all 3

3.5 Heart Disease and Cancer Death Rates

25

groups decline over time with clear mortality disadvantages evident for Black or African-Americans. Relative to whites, African-American males have higher heart disease and malignant neoplasm death rates for the 4 timepoints shown in Table 3.2 beginning in 1980. In addition, Hispanics have the lowest heart disease and malignant neoplasm death rates of the 3 groups in 1990, 2000 and 2014; the 3 timepoints with data available for Hispanics (National Center for Health Statistics 2016). Several conclusions can be drawn from the CDC general population life expectancy and death rate data. First, over time life expectancy has lengthened, and death rates have declined for all 3 groups of interest. Second, during the limited period that CDC data have been available for Hispanics, life expectancy is longer and death rates are lower compared to African-Americans as well as non-Hispanic whites. Finally, life expectancy and death rates are more similar between Hispanics and non-Hispanic whites than they are between African-Americans and non-Hispanic whites.

3.6 The Hispanic Paradox Despite being at the low end on educational attainment and income spectrums, available CDC statistics consistently document superior mortality outcomes among Hispanics living in the US relative to other groups including non-Hispanic whites. This so-called “Hispanic Paradox” has been chronicled in numerous studies since 1980 (Franzini et al. 2001; Ruiz et al. 2013; Markides and Coreil 1986). One of the earliest journal articles documenting what appeared to be an epidemiological mortality paradox for Hispanics relative to whites was published in 1986 by Markides and Coreil (1986). These authors found similarities between Hispanics living in the Southwest US, mainly Mexican-Americans, and non-Hispanic whites on infant mortality, life expectancy, and mortality rates for certain cancers as well as cardiovascular disease. Beginning in the 1990s, additional research was conducted to examine this further and the concept of a Hispanic Paradox evolved. In fact, several of the subsequent studies document outright mortality advantages, rather than similarities, for selected Hispanic populations relative to non-Hispanic whites (Markides et al. 1997; Sorlie et al. 1993). “The evidence in support of the [Hispanic mortality] advantage has been based on a variety of data sources, including vital statistics, community surveys linked to National Death Index data, Medicare data linked to application records for Social Security cards maintained [by] the Social Security Administration …., and mortality follow-ups by individual regional studies” (Markides and Eschbach 2005). In several studies, the Hispanic Paradox is shown to be evident primarily within older age groups (Markides et al. 1997; Hummer et al. 2004). It has also been determined that the paradox is most apparent in immigrant Hispanics (Hummer et al. 2000), especially Mexican immigrants (Palloni and Arias 2004). The paradox is not, however, readily evident among US-born Hispanics (Singh and Hiatt 2006).

26

3 Literature and Statistical Review of Race/Hispanic Origin …

Nevertheless, questions have been raised regarding the Hispanic Paradox. The paradox seems more difficult to establish in research where all deaths within a population have been identified (Smith and Bradshaw 2006). Other potential issues with the Hispanic Paradox have included “…data problems such as a lack of comparability in reporting of Hispanic origin in vital statistics and census records, age misreporting, and difficulties in linking persons of Hispanic origin among various data sources…” (Turra and Elo 2008). Feasible alternative hypotheses have been proposed attempting to explain the apparent mortality advantage of Hispanics living in the US. One explanation that has received considerable attention in the scientific literature is the “unhealthy outmigrant hypothesis” (Pablos-Mendez 1994), commonly referred to as the “salmon bias.” Briefly, this hypothesis postulates that out-migration may be more frequent among US residents born in Latin American countries after their health deteriorates. Such people could return to their native countries for cultural and other reasons, particularly at older ages, and die in their home countries rather than in the US. As a result, US death certificates are not issued for these individuals, and they fail to be counted in US mortality statistics. To the extent that this occurs, this has the effect of underestimating US death rates and overestimating life expectancy among Hispanics who continue to reside in the US. If the salmon bias is a realistic explanation for the Hispanic Paradox, it would logically have its greatest impact among older individuals, possibly in failing health, as well as immigrants from countries more proximate to the US, like Mexico. In fact, these are 2 of the groups, whose mortality outcomes provide the strongest evidence for the paradox (Markides and Eschbach 2005). In brief, while there has been substantial support for the Hispanic Paradox from a variety of sources in the scientific literature, questions exist, and alternative hypotheses cannot be readily ruled-out. Special groups, like former Major League Baseball players have not been studied in relationship to the Hispanic Paradox or the salmon bias. The current study contains too few Hispanics to confidently address questions related to the paradox. Moreover, this study is focused on a very special group of Hispanics, namely former professional baseball players, and their mortality experiences could be dissimilar to their counterparts in the general population. Nevertheless, by studying former Major League Baseball players born in Latin American countries, clues can be provided that might help guide future research efforts related to the study of the Hispanic Paradox.

3.7 Professional Athlete Literature Baseball, basketball and football players have contributed to our understanding of mortality risk among professional athletes based on race, but there is little or no information on this topic as it relates to individuals of Hispanic origin.

3.7 Professional Athlete Literature

27

3.7.1 Baseball Several published studies have examined mortality among former Major League Baseball players (Panjer 1993; Saint Onge et al. 2008; Reynolds and Day 2012). However, comparisons of mortality outcomes based on race, and especially Hispanic origin, are quite limited. Life expectancy was investigated by Saint Onge and colleagues (2008) in their study of 6772 Major League Baseball players who debuted between 1902 and 2004. However, undisclosed data limitations did not allow for life expectancy comparisons based on race/ethnicity.

3.7.2 Basketball Critical literature on mortality and race among professional basketball players comes from 2 publications (Lawler et al. 2012; Markowitz 2018). However, neither of these studies could reasonably examine mortality among Hispanic basketball players. While the number of NBA players born in Latin American countries has increased over time, there are too few Hispanic players to study in a meaningful way. Lawler and colleagues (2012) conducted a 59-year historical cohort study of more than 3300 professional basketball players who played in the US anytime between 1946 and 2005. In all 4 Cox regressions conducted with slightly different sets of variables, there is about a 70–85% increased risk of mortality among African-Americans compared to whites that, in all cases, is statistically significant. Markowitz (2018) compared mortality risk between African-American and white NBA players who played in the League between 1960 and 1986 using multivariate Cox models that control for year of birth, BMI, years of career playing experience, and birthplace region. The results indicate that African-Americans have a statistically significant (86%) increased mortality risk relative to whites. When only significant variables are retained in a second Cox analysis, the mortality risk for African-Americans remains elevated by 58% and continues to be significant (Markowitz 2018).

3.7.3 Football In 1994, researchers from the National Institute for Occupational Safety and Health (NIOSH) reported on the causes, risks and rates of mortality among about 6800 former NFL players who were active between 1959 and 1988 (Baron and Rinsky 1994). The study uses a vital status end date of December 31, 1991 and because of this, the median age of the study cohort is only about 41 years. Predictably, the number of reported deaths, 103 or about 2% of the cohort, is low. Nevertheless, comparisons of mortality to the general population reveal a decreased risk of dying among the NFL players relative to the general population. Despite this advantage for the entire

28

3 Literature and Statistical Review of Race/Hispanic Origin …

NIOSH cohort, linemen have elevated rates of cardiovascular mortality compared to the general population. As a follow-up to this finding, within-player analyses are conducted to identify risk factors for cardiovascular deaths. In “internal analyses,” which divide the cohort into whites and non-whites, there is a 70% significant increased risk of cardiovascular mortality for non-whites in analyses that separately control for BMI and player position (Baron and Rinsky 1994). The NIOSH researchers later conducted a 17-year follow-up study with a subgroup of their 1994 cohort limited to players who had at least 5 years of pensioncredited service in the NFL (Baron et al. 2012). Internal risk factor analyses of cardiovascular mortality risk examine the effect of BMI, era of play, player position and race. When race is entered in 3 of the 4 Cox regressions, hazard ratios predicting cardiovascular mortality risk ranges from 1.57 to 1.71 for non-whites, and in all 3 instances, is statistically significant (Baron et al. 2012). Markowitz (2018) studied all-cause mortality risk factors among former NFL players who played in the league anytime between 1960 and 1986. Just over 6500 US-born players have non-missing race data and are included in the analyses of race. In a Cox regression that control for year of birth, African-American NFL players have a 59% increased risk of dying compared to white players. In multivariate analyses that control for year of birth plus all other significant predictors of all-cause mortality, African-American race is a significant independent risk factor for mortality with a hazard ratio that reflects a 30% increased risk of dying relative to whites (Markowitz 2018). Attention will now be turned to the empirical study of former Major League Baseball players that is the focus of this book. The first step will be to delineate the methods and specify the hypotheses. This will be followed by preliminary statistical analyses, and then increasingly more sophisticated analyses that consider other potentially important variables.

References Arias, E., Heron, M., & Hakes, J. K. (2016). The validity of race and Hispanic-origin reporting on death certificates in the United States: An update. Vital and Health Statistics, Series, 2, 172(172), 1–23. Baron, S. L., & Rinsky, R. (1994). Health hazard evaluation report, National Football League players mortality study. Report No. HETA 88-085. Atlanta: Centers for Disease Control and Prevention/National Institute for Occupational Safety and Health. Retrieved from https://www. cdc.gov/niosh/hhe/reports/pdfs/1988-0085-letter.pdf?id=10.26616/NIOSHHETA88085. Baron, S. L., Hein, M. J., Lehman, E., & Gersic, C. M. (2012). Body mass index, playing position, race, and the cardiovascular mortality of retired professional football players. American Journal of Cardiology, 109, 889–896. Brooks, E. G., & Reed, K. D. (2015). Principles and pitfalls: A guide to death certification. Clinical Medicine and Research, 13(2), 74–82. CDC (2018). Leading causes of death in males, 2015. Retrieved from https://www.cdc.gov/ healthequity/lcod/men/2015/race-ethnicity/index.htm.

References

29

Franzini, L., Ribble, J. C., & Keddie, A. M. (2001). Understanding the Hispanic paradox. Ethnicity and Disease, 11, 496–518. Hummer, R. A., Rogers, R. G., Amir, S. H., Forbes, D., & Frisbie, W. P. (2000). Adult mortality differentials among Hispanic subgroups and non-Hispanic Whites. Social Science Quarterly, 81, 459–476. Hummer, R. A., Benjamins, M. R., & Rogers, R. G. (2004). Racial and ethnic disparities in health and mortality among the U.S. elderly population. In R. A. Bulatao & N. B. Anderson (Eds.), Understanding racial and ethnic differences in health in late life: A research agenda. Washington: National Academy Press. Lawler, T., Lawler, F., Gibson, J., & Murray, R. (2012). Does the African-American-white mortality gap persist after playing professional basketball? A 50-year historical cohort study. Annals of Epidemiology, 22, 406–412. Markides, K. S., & Coreil, J. (1986). The health of Southwestern Hispanics: An epidemiologic paradox. Public Health Reports, 101, 253–265. Markides, K. S., & Eschbach, K. (2005). Aging, migration, and mortality: Current status of research on the Hispanic Paradox. The Journals of Gerontology, Series B, 60(2), 68–75. Markides, K. S., Rudkin, L., Angel, R. J., & Espino, D. V. (1997). Health status of Hispanic elderly in the United States. In L. G. Martin & B. Soldo (Eds.), Racial and ethnic differences in the health of older Americans. Washington: National Academy Press. Markowitz, J. S. (2018). Mortality and its risk factors among professional athletes: A comparison between former NBA and NFL players. Cham: Springer. National Center for Health Statistics. (2016). Health, United States, 2015: With special feature on racial and ethnic health disparities. Washington, D.C.: US Government Printing Office. Pablos-Mendez, A. (1994). Mortality among Hispanics. JAMA, 271(16), 1237. (letter; comment on: JAMA, 1993, 270(20), 2464–2468). Palloni, A., & Arias, E. (2004). Paradox lost: Explaining the Hispanic adult mortality advantage. Demography, 4, 385–415. Panjer, H. H. (1993). Mortality differences by handedness: Survival analysis for a right-truncated sample of baseball players. Transactions of Society of Actuaries, 45, 257–274. Reynolds, R. & Day, S. (2012). Life expectancy and comparative mortality of Major League Baseball players, 1900–1999. WebmedCentral Sports Medicine, 3(5). Retrieved from http://www. webmedcentral.com/article_view/3380. Ruiz, J. M., Steffen, P., & Smith, Y. B. (2013). Hispanic mortality paradox: A systematic review and meta-analysis of the longitudinal literature. American Journal of Public Health. Retrieved from https://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2012.301103. Saint Onge, J. M., Rogers, R. G., & Krueger, P. M. (2008). MLB players’ life expectancies. Social Science Quarterly, 89(3), 817–830. Singh, G. K., & Hiatt, R. A. (2006). Trends and disparities in socioeconomic and behavioural characteristics, life expectancy, and cause-specific mortality of native-born and foreign-born populations in the United States, 1979–2003. International Journal of Epidemiology, 35(4), 903–919. Smith, D. P., & Bradshaw, B. S. (2006). Rethinking the Hispanic Paradox: Death rates and life expectancy for US non-Hispanic white and Hispanic populations. American Journal of Public Health, 96(9), 1686–1692. Sorlie, P. D., Backlund, M. S., Johnson, J. K., & Rogat, F. (1993). Mortality by Hispanic status in the United States. JAMA, 270, 2646–2648. Turra, C. M., & Elo, I. T. (2008). The impact of Salmon Bias on the Hispanic mortality advantage. Population Research and Policy Review, 27, 515–530.

Part II

Empirical Study Description and Preliminary Results

Chapter 4

Study Methods

4.1 Study Objective and Design This is a retrospective cohort study based on player-specific and baseball-related data. The primary objective of this study is to determine whether there are differences in all-cause mortality risk among 3 groups of former Major League Baseball players, i.e., African-Americans, Hispanics, and non-Hispanic whites, who participated in the League during the 40 years following the end of segregation in baseball beginning in 1947. The study cohort includes all individuals who played Major League Baseball anytime between 1947 and 1986. These years were selected for several reasons. First, Major League Baseball’s color barrier was broken in 1947. Not only did this open the door to a slow but steady influx of African-American players into Major League Baseball, but dark-skinned Hispanics, who previously had not been welcomed into the League, were then permitted to play as well (Stewart and Kennedy 2002). Only a small percentage of players who played Major League Baseball after 1986 will be deceased by the beginning of 2018. Given the costs and efforts to collect data on additional (younger) players, the large majority of whom are still alive, the decision was made to end data collection with individuals who played in 1986. In retrospective database studies, it is sometimes difficult or impossible to obtain data on some variables of interest. In an ideal world, a mortality study would include variables likely to impact on illness and well-being, such as medical conditions, cigarette smoking, alcohol and drug use, exercise, nutrition, and family history. However, when the research is based on historical records that go back decades, the available variables may be limited. Although this is the case with respect to the current study, the variables considered in this book do allow for a reasonable test of the study hypotheses and should add significantly to existing knowledge in this area.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2019 J. S. Markowitz, Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball, SpringerBriefs in Public Health, https://doi.org/10.1007/978-3-030-17280-0_4

33

34

4 Study Methods

4.2 Data Sources This study uses retrospective data from a variety of sources. The main data source is the Baseball Reference web-site (Sports Reference LLC 2018) which is one of the leading Internet sports statistics sites in the world. According to the Society for American Baseball Research, known as SABR, “[Baseball Reference.com is] easily the best source for precalculated historical statistics” (SABR a no date). In fact, Baseball Reference essentially offers a complete statistical history of every Major League Baseball team and all players. Players in the study cohort were identified by creating a database with the names and birth dates of every individual on a Major League Baseball team between 1947 and 1986. This information was obtained from the Baseball Reference web-site by clicking on each team listed in the “Teams” section, then “Franchise History,” and subsequently each season beginning with 1947 and continuing through 1986 (Sports Reference LLC 2018). This same procedure was repeated for each Major League Baseball team in existence between 1947 and 1986. A Statistical Analysis System (SAS) program (SAS 2018) was written to extract all unique players based on their first and last names as well as their dates of birth.

4.3 Hypotheses Based on existing general population and professional athlete mortality literature, it is hypothesized that mortality risk will be significantly elevated among former AfricanAmerican Major League Baseball players compared to Hispanic and non-Hispanic white players. No significant difference in mortality risk is predicted between Hispanic and non-Hispanic white players. These hypothesized relationships are expected to persist after controlling for year of birth and other potentially important variables. Existing scientific literature provides a strong justification for expecting AfricanAmericans to be the most disadvantaged group when it comes to mortality (see Chap. 3). While Hispanics living in the US fare better mortality-wise than nonHispanic whites in vital statistics, the general population literature is less consistent. Moreover, there are no mortality comparisons in the literature between Hispanics and non-Hispanic baseball players; accordingly, the null hypothesis is predicted between these groups.

4.4 The Primary Independent Variable: Race/Hispanic Origin As the main independent variable for this study, race/Hispanic origin is comprised of 3 groups:

4.4 The Primary Independent Variable: Race/Hispanic Origin

35

• non-Hispanic African-Americans (hereafter called African-Americans) • Hispanics • non-Hispanic whites. Hispanic in this study is defined as individuals born in Latin American countries or Puerto Rico. Birthplace information, including country of birth, is publicly available on the “Players” pages of Baseball Reference (Sports Reference LLC 2018). The race of American-born players was captured using a multistep approach. Unlike variables, such as players’ weight, height and date of birth, race is generally not a publicly available variable. Therefore, race was obtained using a variety of sources, methods and techniques that include: • the book titled “A hard road to glory: Baseball” (Ashe Jr. 1993) • select historical baseball facts • pictures mainly from baseball collector cards. Prior to his death in 1993, Arthur Ashe Jr. the American tennis star, wrote a series of books on the Black athlete, including one specifically on baseball (Ashe Jr. 1993). In each of these books, Ashe Jr. includes extensive lists of African-American athletes. These lists, however, are not exhaustive and have to be supplemented with other methods and sources. In Major League Baseball, many historical facts exist with respect to players’ race. A well-referenced Wikipedia piece called “List of first black Major League Baseball players” identifies the first African-American players for each Major League Baseball team as well as their start dates in the League (Wikipedia 2018). For example, in 1959, the Boston Red Sox was the last Major League Baseball team to sign an AfricanAmerican player. The Detroit Tigers fielded their first African-American in 1958 and the Philadelphia Phillies did so in 1957. It can be safely concluded that individuals on these teams who played prior to these respective seasons were either white or born in Latin American countries. Several players born outside of the US and Latin America were dropped and excluded from all statistical analyses. For purposes of this study, individuals born in a Latin American country would automatically be placed into the Hispanic group of players. Based on the respective years that the other Major League Baseball teams fielded their first African-American player, the same logic is used to identify the race of hundreds of additional players. Several other race-related historical baseball facts further facilitated the identification of race for dozens of additional African-American and non-Hispanic white players. For example, there were several all-Black colleges, and players who attended these schools were all coded as African-American. In many cases, the historical facts serve to confirm race for individuals who have already been identified via the Ashe Jr. book and/or player photos (described below). More so than any other professional sport, baseball is known for its long and storied tradition of collector cards. The first card goes back to the 1860s when they appeared in packs of cigarettes as a means of increasing sales. These rectangular cardboard cards later appeared in the twentieth century in packages accompanied by a piece of bubble gum and were manufactured by numerous companies. On the front

36

4 Study Methods

of most of these cards is a photo of a Major League Baseball player. The likely race of most players can be determined from these collector cards, which now appear on Internet sites including eBay, Amazon, Wal-Mart, Target and dozens of other sports memorabilia sites. Finally, Googling the names of former Major League Baseball players can yield hundreds of images of the players of interest. Because race is essentially a subjective construct, there is no guarantee that a player who appears to be white or African-American on a collector card or an Internet image, identifies himself as being that race. There is also the possibility that specific players come from mixed racial and ethnic backgrounds. The fact is that any player can be misclassified based on a photo alone. Although the multistep approach used to identify player race should minimize error, complete accuracy is uncertain. In this study, it can be assumed that every Hispanic player is classified correctly since this category is based solely on published country of birth.

4.5 The Primary Dependent Variable: Time to All-Cause Mortality Using a vital status end-date of January 1, 2018, the main dependent variable for this study is time to all-cause mortality, or simply all-cause mortality. Cause-specific mortality data on former Major League Baseball players are not available to this author. All players who died on or before January 1, 2018 will be classified as dead, and all others will be considered alive. Some players classified “alive” for purpose of this study, have died sometime between January 2, 2018 and the writing of this book. This is a small limitation, but without a real-time means of capturing player deaths, no other options are available. Vital status is a simple alive/dead variable that does not consider age at death for those who have died, or the age of players still alive. Admittedly crude, vital status begins to be meaningful when year of birth is simultaneously considered. Hence, analyses of vital status will be stratified by year of birth categories (to be detailed below). Time to death, or age on the designated end-date for players still alive (January 1, 2018), is a so-called “time-to-event” variable that will be analyzed using Cox proportional hazards regressions, or Cox regressions or models (Cox and Oakes 1984). This survival procedure can be used with multiple independent variables and covariates and estimates the time it takes to experience a specific event. For purposes of the current study, time to death is the focus. Players who have died are considered “uncensored” observations, while those still alive on January 1, 2018 are “censored.” Age at death is known with respect to uncensored players and is considered in the Cox models. What is known about censored players is that they have “survived” to the end-date without experiencing death. For purposes of the survival analyses contained in this book, the age of censored players on January 1, 2018 is the critical data piece. In Cox proportional hazards regressions, individuals who are censored

4.5 The Primary Dependent Variable: Time to All-Cause Mortality

37

contribute to the number of players who are at risk of dying based on their age on the vital status end-date (Allison 2010). In addition to baseball demographic and player-related information, such as height, weight, years of play, player position and handedness, the “Players” pages of Baseball Reference also provide vital status data that, when applicable, includes date of death (Sports Reference LLC 2018). Data obtained from these sites have been used successfully in several prior mortality research studies of professional athletes (Lawler et al. 2012; Markowitz 2016; Markowitz 2018). The mortality data published on the Baseball Reference Internet pages has been double-checked with vital status data available at other sites and in other databases. For example, the Lahman Baseball database (2017) contains comprehensive Major League Baseball player death information in its “Master File.” This database has been characterized by SABR as “… basically a standard Baseball Encyclopedia in downloadable form” (SABR a no date). As a test of the validity of the mortality data found on the “Players” pages of the Baseball Reference Internet-site, vital status was compared with the death data found in the Lahman database. At the time that this validity test was conducted, the Lahman database was updated through the end of 2016. The vital status of 100 randomly selected players in the study cohort who were still alive plus 100 randomly selected players who had passed away (through the end of 2016) according to Baseball Reference were compared with the data compiled by Lahman. In all 200 instances, the vital status of the players matches between these two databases. In addition, the dates of death match with respect to the players who have passed away. These results increase confidence, although do not ensure, that the mortality information contained at the Baseball Reference site is accurate and complete. Former Major League Baseball players often remain in the media spotlight for years after retiring from baseball and it is common for their death information to be published on additional Internet and non-Internet sites. While there are many published sources of such material, these tend not to be systematic or complete. Nevertheless, death information on former Major League Baseball players can be found at numerous sports and celebrity Internet-sites as well as more general publications that include the Baseball Almanac (2018), Sporting News (2017), Tributes (2017), Find-a-Grave.com (2017), and the Los Angeles Times (2017). In addition to the validity checks detailed above utilizing the Lahman database, additional checks were done to compare death data found at the Baseball Reference site with other mortality information sources listed in this paragraph and, again, no discrepancies were found. These additional checks focused on players who died at relatively young or old ages, i.e., under 40 or over 90 years, since these may be the most prone to error.

38

4 Study Methods

4.6 The Primary Covariate: Year of Birth Mortality and age are statistically linked. Beginning around the early teen years, the probability of dying increases with each successive year (Social Security Administration no date). Hence, age must be viewed as an essential covariate in mortality studies. In this study, year of birth will serve as the main covariate. While date of birth would be more accurate than birth year, analyzing year of birth provides more comprehensible interpretations of the data. Year of birth will be used in the analysis as both a categorical and continuous variable, depending on the analysis being conducted. Continuous year of birth will be slightly more accurate, but again, its categorical counterpart may improve interpretability. The specific year of birth categories to be used in this study are as follows: • • • • • •

1905–1914 1915–1924 1925–1934 1935–1944 1945–1954 1955–1966. Two players in the study cohort were born in 1966.

4.7 Other Variables In addition to race/Hispanic origin, (the main independent variable), year of birth, (the primary covariate), and all-cause mortality, (the primary dependent variable), several additional items will be analyzed in this book. This will include educational attainment, body mass index (BMI), US birthplace region, years of Major League Baseball playing experience, playing position and handedness. Conceptually, these are all independent variables that may add to the prediction of mortality risk above and beyond what is already explained by race/Hispanic origin. Alternatively, these other variables could help to clarify any relationship that is uncovered between race/Hispanic origin and all-cause mortality. Each of these other variables will now be described.

4.7.1 Educational Attainment Two independent sources are used to measure educational attainment in this study as follows: • Player pages of Baseball Reference (Sports Reference LLC 2018) • College Playing spreadsheet of the Lahman Baseball Database (2017).

4.7 Other Variables

39

Baseball Reference (Sports Reference LLC 2018) provides the names of the high schools and colleges attended by each player but does not indicate the number of years completed or whether the respective players graduated. The Lahman Baseball Database (2017) gives the number of years of college attendance for the sub-group of players who attended college in their “College Playing” database. In the rare instances where there was a discrepancy between these two educational attainment data sources, individual SABR biographies (SABR b no date) were consulted, if available. Based on these information sources, the following 5 categories of educational attainment have been created: • • • • •

little or no high school at least some high school, but no college some college, but unknown number of years completed 1–2 years of college completed 3 or more years of college completed.

These categories are essentially ordinal, rather than continuous, and will be treated as categorical items in the analysis.

4.7.2 Body Mass Index (BMI) BMI is a measure of weight that is adjusted for height. Playing-time BMI during the last year of play in Major League Baseball will be used in this study. Height and weight data have beem obtained from the Players pages of Baseball Reference (Sports Reference LLC 2018). According to the World Health Organization, (W.H.O. 2008), BMI can be aggregated into 4 categories: • • • •

underweight (less than 18.5) normal (between 18.5 and 24.99) overweight (between 25 and 29.99) obese (30 and over).

There are further sub-classifications of obese BMIs that do not apply to the study cohort. Also, none of the players in the study cohort are in the underweight BMI category and only 3 players are obese. When BMI is analyzed as a categorical variable, these 3 players will be aggregated with the overweight group. Therefore, only 2 of the 4 BMI categories, normal and overweight, will be utilized in this study. The BMI categories are clinically meaningful, but continuous BMI may be more exact as no information is lost. Depending on the analysis, categorical and/or continuous BMI will be used.

40

4 Study Methods

4.7.3 Number of Career Years Played in Major League Baseball A player is credited as participating in 1 year of Major League Baseball for each regular season that he played 1 or more games. Number of career years played Major League Baseball is a continuous variable. However, to make the results more comprehensible, ordered categories of this variable will also be created and used in the analysis as follows: • • • • •

1–2 years 3–5 years 6–9 years 10–14 years 15 years or more.

4.7.4 Player Position Player position is aggregated for the analysis in 2 ways. First, a dichotomous classification of pitchers and non-pitchers is created. Second, players are placed into one of the following 5 groups: • • • • •

catchers pitchers infielders outfielders other.

Infielders encompass first, second and third basemen, plus shortstops. Outfielders include right-, left- and center- fielders. The “other” position category includes designated hitters, pinch-runners, and pinch-hitters. When individuals play more than a single position, the position played in the most games is used in the analyses.

4.7.5 US Birthplace Region Four US regions, i.e., Northeast, Midwest, South and West, are designated by the Census (US Census Bureau 2015) and are used by the CDC and other government agencies and organizations. Figure 4.1 contains a map of the US with each of the 4 regions color-coded. Birthplace region, or the region of the US where players are born, cannot be fully examined when Hispanics are included in the analyses. This is because all Hispanics in this study are, by definition, born in Latin American countries.

4.7 Other Variables

41

Midwest

Northeast

(12 states ) WA MT OR

VT MN

WI

SD

ID WY NV UT

CD

IL KS

NY

MI

IA

NE

CA

(9 states)

ND

IN

PA

OH

MO

WV

VA

KY AZ

NM

OK

ME NH MA RI CT NJ DE MD DC

NC

TN SC

AR MS

AL

GA

LA AK

HI

TX

West (13 states)

FL

South (17 states including DC)

Fig. 4.1 Census regions and divisions of the United States

4.7.6 Handedness In baseball, handedness can be defined in terms of the hand used to throw and the side used to bat. A player can bat from the right or left side of home plate and can also switch-hit, i.e., bat from either side of home plate. Hence, two handedness variables are formed based on throwing arm and batting side respectively.

4.8 Statistical Plan This section details the statistical plan that will be used to describe the study cohort and to test the main study hypotheses. The analytic strategy will also include systematic testing of other independent variables. The empirical analysis will begin in Chap. 5 by describing the study cohort and examining the relationship between race/Hispanic origin and variables related to births and deaths, specifically year of birth and vital status. In addition, separate cross-tabulations of race/Hispanic origin by the year of birth categories and vital status will be accompanied by chi-square tests. Chap. 5 will conclude with a Cox proportional hazards regression that examines the year of birth categories in relation to all-cause mortality. One of the key statistics derived from Cox regressions is the hazard ratio. This statistic reflects the risk of experiencing the event of interest, i.e., all-cause mortality, for the 3 groups of players, and can also contain other independent variables

42

4 Study Methods

and covariates. Index referencing is a procedure that can be used in Cox models to determine mortality risk (based on hazard ratios) relative to a given group. As a hypothetical example, suppose overweight BMI players are being compared to normal weight individuals on mortality risk. The normal weight players could be assigned as the index reference and given a value of 0, while the overweight players would be given a value of 1. In this example, the hazard ratio would provide the estimated risk of dying for the overweight group using the normal weight group as the reference. For readers familiar with odds ratios, hazard ratios can be interpreted in a similar manner. For example, a hazard ratio of 1.75 in the hypothetical example, would indicate that overweight players have a 75% increased risk of all-cause mortality relative to normal BMI players. A hazard ratio that is less than 1 denotes a decreased risk of the outcome of interest. Continuing with the hypothetical example, a hazard ratio of 0.6 would reflect a 40% decreased risk of all-cause mortality for the overweight BMI group relative to those with a normal BMI. The testing of the main study hypotheses related to mortality risk among the 3 race/Hispanic origin groups of players will be accomplished using a multi-level approach beginning with preliminary tests provided in Chap. 6. Comparisons will show vital status by race/Hispanic origin for all year of birth categories. Chapter 6 will conclude with a Cox model that tests whether mortality risk varies by race/ Hispanic origin controlling for year of birth. Assuming a statistical relationship between race/Hispanic origin and mortality (controlling for year of birth), the next step in the analytic process will be to examine additional independent variables to determine whether they add to the prediction of mortality risk. Alternatively, these other variables could be distorting relationships between race/Hispanic origin and mortality. This examination will engender at least 2 tests of significance to determine whether there is a relationship between each of the additional independent variables and: 1. race/Hispanic origin (using chi-square tests) and 2. mortality risk (using Cox models). This latter test will control for year of birth and assess the relationship between each of these variables and all-cause mortality. If significant relationships are uncovered in both tests, then race/Hispanic origin will be added to the Cox model above. In the final, most definitive testing of the study hypotheses, all the significant variables will be included along with race/Hispanic origin and year of birth in a multivariate Cox proportional hazards regression (Chap. 9). This test will indicate whether race/Hispanic origin persists as a predictor of mortality when all the significant study variables are controlled simultaneously. The analyses described thus far are “internal” since they compare different groups of players within the study cohort. Chap. 10 will describe “external” analyses that separately compare mortality rates of players within the 3 race/Hispanic origin groups with gender- and, age-matched controls from the general population. Details on the statistical plan for the external analyses will be provided separately in Chap. 10. Many statistical tests will be conducted in this book using an alpha level of 5% and no correction to this level will be made. Because of this, some false positive findings are likely. Readers are cautioned about this possibility.

4.9 SAS Procedures

43

4.9 SAS Procedures The data analysis for this book will use SAS/BASE and SAS/STAT software, Version 9.4 of the SAS System for Windows 10 (SAS Institute Inc. 2018). The specific SAS procedures to be used include: Proc FREQ for frequency distributions and cross tabulations, Proc UNIVARIATE for descriptive statistics and histograms, Proc ANOVA for analysis of variance, Proc GLM for analysis of covariance, Proc PHREG for Cox proportional hazards regressions, and Proc STDRATE to obtain Standardized Mortality Ratios (SAS Institute Inc. 2018).

References Allison, P. D. (2010). Survival analysis using SAS: A practical guide (2nd ed.). Cary: SAS Institute Inc. Ashe, A., Jr. (1993). A hard road to glory: The African-American athlete in basketball. New York: Amistad Press. Baseball Almanac. (2018). MLB players by year of death. Retrieved from http://www.baseballalmanac.com/players/baseball_deaths.php. Cox, D. R., & Oakes, D. (1984). Analysis of survival data. New York: Chapman & Hall. Find-a-Grave.com (2017). Retrieved from https://www.findagrave.com. Lahman Baseball Database. (2017). College playing spreadsheet. Retrieved from http://www. seanlahman.com/baseball-archive/statistics/. Lawler, T., Lawler, F., Gibson, J., & Murray, R. (2012). Does the African-American-white mortality gap persist after playing professional basketball? A 50-year historical cohort study. Annals of Epidemiology, 22, 406–412. Los Angeles Times. (2017). Obituaries database. Retrieved from http://obituaries.latimes.com/ category/sports/1/. Markowitz, J. S. (2016). Lost seasons: Arrests, suspensions, career chaos and mortality among National Football League (NFL) players. North Charleston: CreateSpace. Markowitz, J. S. (2018). Mortality and its risk factors among professional athletes: A comparison between former NBA and NFL players. Cham: Springer Nature. SABR a, Society for American Baseball Research, (no date). A guide to sabermetric research: How to find raw data. Retrieved from https://sabr.org/sabermetrics/data. SABR b, Society for American Baseball Research, (no date). SABR baseball biography project. Retrieved from http://sabr.org/bioproject. SAS Institute Inc. (2018). SAS/BASE and SAS/STATS, Version 9.4. Cary: SAS Institute Inc. Social Security Administration. (no date). Actuarial Life Table. Table entitled “Period Life Table, 2014”. Retrieved from https://www.ssa.gov/oact/STATS/table4c6.html. Sporting News. (2017). Notable sports deaths of 2016. Retrieved from http://www.sportingnews. com/other-sports/photos/notable-sports-deaths-of-2016-nfl-football-nba-basketball-baseball/ e5dyl44gj40h1hbeq6bu9pm0o. Sports Reference LLC. (2018). Players pages. Retrieved from https://www.baseball-reference.com/ leagues/MLB/1960-opening-day-rosters.shtml. Stewart, M., & Kennedy, M. (2002). Hispanic baseball’s finest fielders. Brookfield: Twenty-First Century Books. Tributes. (2017). Celebrity deaths in sports. Retrieved from http://www.tributes.com/celebrity/ deaths/Sports.

44

4 Study Methods

US Census Bureau. (2015). Geographic terms and concepts—Census divisions and census regions. Retrieved from https://www.census.gov/geo/reference/gtc/gtc_census_divreg.html. Wikipedia. (2018). List of first Major League Baseball players. Retrieved from https://en.wikipedia. org/wiki/List_of_first_black_Major_League_Baseball_players. W.H.O. (World Health Organization). (2008). BMI classification. Retrieved from http://apps.who. int/bmi/index.jsp?introPage=intro_3.html.

Chapter 5

Births and Deaths in the Study Cohort

5.1 Results 5.1.1 Cohort Description A total of 5263 former players who played Major League Baseball anytime between 1947 and 1986 are included in the analysis consisting of 4258 non-Hispanic whites (80.9%), 587 African-Americans (11.2%), and 418 Hispanics (7.9%). A total of 2018 of these 5263 former players (38.3%) died, while 3245 (61.7%) were still alive as of January 1, 2018. The mean and median year birth for the entire cohort is 1941.1 and 1943 respectively. This indicates that players still alive on the vital status end-date of January 1, 2018 average about 77–79 years of age.

5.1.2 Year of Birth by Race/Hispanic Origin See Table 5.1 for a description of race/Hispanic origin and both the categorical and continuous versions of year of birth. The percentages of players born within each year of birth category are displayed, followed by mean and median year of birth for the 3 race/Hispanic origin groups. Clearly, non-Hispanic whites have an earlier year of birth compared to African-American and Hispanic players. This is confirmed by the significant overall test results for the comparisons of race/Hispanic origin for both the categorical and continuous versions of year of birth (p’s < 0.0001). There are much higher percentages of non-Hispanic white players born in the earliest year of birth categories. For example, nearly 1 in 5 (18.2%) non-Hispanic white players are born prior to 1925 compared to less than 4% of African-Americans and 6.3% of Hispanics. For the more recent year of birth categories, the reverse pattern is evident; more than two-thirds of African-American players (67%) are born between 1945 and 1966 compared to 56.7% of Hispanics and only 42.5% of non-Hispanic © The Author(s), under exclusive license to Springer Nature Switzerland AG 2019 J. S. Markowitz, Mortality Among Hispanic and African-American Players After Desegregation in Major League Baseball, SpringerBriefs in Public Health, https://doi.org/10.1007/978-3-030-17280-0_5

45

46

5 Births and Deaths in the Study Cohort

Table 5.1 Year of birth by race/Hispanic origin Year of birth

Race/Hispanic origin AfricanAmerican N

Hispanic %

N

Non-Hispanic white %

1905–1914

2

0.3

6

1915–1924

21

3.6

1925–1934

58

9.9

1935–1944

113

1945–1954 1955–1966

N

%

1.4

122

2.9

20

4.8

655

15.4

59

14.1

818

19.2

19.3

96

23.0

855

20.1

207

35.3

114

27.3

1010

23.7

186

31.7

123

29.4

798

18.7

Totalsa

587

100.1

418

100.0

4258

100.0

Mean/median year of birth

1947.8/1950

1945.6/1948

pvalue*

1939.7/1941